Dietary Methionine Requirements for Juvenile Florida Pompano (Trachinotus carolinus)

Abstract

A 56-day feeding trial was conducted to evaluate the quantitative methionine requirements in the diets of Florida pompano (Trachinotus carolinus). Eight practical diets using soybean meal, poultry meal, and red lentil meal as the primary protein sources were formulated using graded levels of methionine supplement (0 to 0.70 g/100 g diet). Groups of 15 juvenile Florida pompano (4.04 ± 0.05 g) were size-sorted and placed into one of 40 glass aquaria (132 L) with five replicates per diet. Significant differences (p ≤ 0.05) were observed in overall biomass, mean weight, weight gain, thermal growth coefficient (TGC), and feed conversion ratio (FCR). To estimate the dietary methionine requirement, a series of statistical models, including the one-slope broken line model (BLM1), two-slope broken line model (BLM2), broken quadratic model (BQM), and four-parameter saturation kinetic model (SKM-4) were used to assess mean weight, weight gain, TGC, apparent net protein retention (ANPR), and methionine retention (MR). The model selection showed that BLM1 fit the data best for MW and TGC, SKM-4 for PWG and ANPR, and BQM for MR. Based on these results, a minimum dietary methionine requirement of 0.68% of the diet or 1.70 g/100 g protein is recommended.

1. Introduction

Historically, fishmeal has been the preferred protein source for commercial diets in aquaculture [1]. This is due to fishmeal possessing most of the necessary nutritional qualities fish need to grow and develop [1,2]. However, utilizing fishmeal in commercial diets has come under scrutiny for a range of reasons, including overfishing of wild stocks, high cost, and limited supply [3]. Because of this, the industry has moved, and will continue to move, towards the use of other protein sources that are more widely available in aquaculture diets. Plant-based proteins have become increasingly popular as a primary protein source in aquaculture diets due to their relative abundance and high protein content [4,5,6].

Legumes present themselves as a good alternative to fishmeal as they have some of the highest protein levels for plant-based ingredients and comprise the most abundant group of plant-based proteins [7,8]. However, several drawbacks are associated with the utilization of plant proteins in commercial aquaculture diets [9]. One main drawback is the nutritional limitations in plant proteins as opposed to fishmeal [9]. Commonly used ingredients such as soybean meal do not meet the necessary levels of several essential amino acids (EAAs) that allow for optimal fish growth and development [10,11]. Essential amino acids (EAAs), such as lysine, methionine, and threonine, are typically the amino acids in shortest supply in the synthesis of proteins, peptides, and other metabolites [11,12,13,14]. Amino acids in the shortest supply of protein synthesis are referred to as limiting amino acids [15]. Methionine is typically seen as one of the first limiting amino acids in fishmeal-free diets utilizing legumes, such as soybeans, as the primary protein source [11,12,16,17,18].

Within the body, methionine is responsible for many important biological functions, including assistance in the regulation of immunity, metabolism, and reproduction [19]. If the dietary methionine requirements are not met, the growth and health of cultured fish may be impeded, and in more serious cases, death may result [20,21,22]. Because methionine requirements vary depending on species, age, and levels of other key amino acids such as cysteine and taurine, it is paramount to understand the dietary methionine requirements of commercially relevant species of interest [18].

Florida pompano (Trachinotus carolinus) is a species of considerable interest within the Americas for domestic aquaculture production [23]. It is a species that is known to accept and grow well on diets composed of plant-based proteins. There has been considerable work done on nutritional requirements of the Florida pompano in hopes of improving their commercial diet [24,25,26]. However, the Florida pompano’s EAA requirements are still not fully understood. This is especially true regarding methionine requirements in Florida pompano, with the only existing works being a thesis that did not give a definitive dietary requirement, as well as an unpublished abstract [12,27]. Therefore, the primary objective of this study is to determine the methionine requirements for the Florida pompano.

2. Material & Methods

2.1. Experimental Diets

Eight experimental diets were formulated and produced by the Laboratory of Aquatic Animal Nutrition (School of Fisheries, Aquaculture, and Aquatic Sciences, Auburn University, Auburn, AL, USA). All diets were formulated to be iso-nitrogenous and isolipidic at 40% protein and 8% lipid. Protein sources for the diets consisted of 12% poultry by product meal, 38.2% solvent-extracted soybean meal, and 22% red lentil meal. Red lentil meal was chosen as a supplemental protein source due to its high protein content and favorable amino acid profile, while being very low in methionine. The source of lipids for the diets was menhaden fish oil at 5.46%. The basal diet (0.57% methionine) was formulated to be deficient in methionine. Diets two through eight were supplemented with increasing levels of crystalline methionine at 0.05%, 0.1%, 0.2%, 0.3%, 0.4%, 0.5%, and 0.7% of the experimental diet (

Table 1. Dietary composition and proximate composition of the eight experimental diets offered to Florida pompano with varying methionine levels.

Composition (As Is)	0.57% Met	0.66% Met	0.71% Met	0.82% Met	0.86% Met	0.94% Met	1.03% Met	1.28% Met
Poultry meal 1	12.00	12.00	12.00	12.00	12.00	12.00	12.00	12.00
Soybean meal 2	38.20	38.20	38.20	38.20	38.20	38.20	38.20	38.20
Lentil meal 3	22.00	22.00	22.00	22.00	22.00	22.00	22.00	22.00
Menhaden fish oil 4	5.46	5.46	5.46	5.46	5.46	5.46	5.46	5.46
Lecithin 5	0.50	0.50	0.50	0.50	0.50	0.50	0.50	0.50
Corn Starch	1.59	1.59	1.59	1.59	1.59	1.59	1.59	1.59
Whole wheat	16.50	16.50	16.50	16.50	16.50	16.50	16.50	16.50
Mineral premix 6	0.25	0.25	0.25	0.25	0.25	0.25	0.25	0.25
Vitamin premix 7	0.50	0.50	0.50	0.50	0.50	0.50	0.50	0.50
Choline chloride 8	0.20	0.20	0.20	0.20	0.20	0.20	0.20	0.20
Rovimix Stay-C 35% 9	0.10	0.10	0.10	0.10	0.10	0.10	0.10	0.10
CaP-dibasic 10	1.50	1.50	1.50	1.50	1.50	1.50	1.50	1.50
Methionine 8	0.00	0.05	0.10	0.20	0.30	0.40	0.50	0.70
Glycin 11	0.70	0.65	0.60	0.50	0.40	0.30	0.20	0.00
Taurine	0.50	0.50	0.50	0.50	0.50	0.50	0.50	0.50
Proximate Analysis %
Crude protein	41.35	40.81	40.75	39.66	40.41	39.00	40.75	40.12
Moisture	8.67	8.38	8.64	11.33	10.65	11.90	8.66	8.48
Crude Fat	7.23	7.36	7.05	6.99	7.16	7.32	7.71	7.94
Crude Fiber	2.71	2.56	2.59	2.69	2.55	2.31	2.10	2.73
Ash	7.44	7.17	7.43	7.31	6.43	6.35	6.62	6.53

¹ River Valley Ingredients, Hanceville, AL, USA. ² De-hulled, solvent extracted soybean meal, Bunge, St. Louis, MO, USA. ³ AGT Food, Minot, ND, USA. ⁴ Omega Protein Inc., Houston, TX, USA. ⁵ The Solae Company, St. Louis, MO, USA. ⁶ Mineral Premix (g 100 g⁻¹ premix): cobalt chloride, 0.004; cupric sulphate pentahydrate, 0.250; ferrous sulfate heptahydrate, 4.0; manganous sulfate anhydrous, 0.650; potassium iodide, 0.067; sodium selenite, 0.010; zinc sulfate heptahydrate, 13.193; and α cellulose, 81.826. ⁷ Vitamin Premix (g/kg Premix): thiamin HCL, 0.5; riboflavin, 8.0; pyridoxine HCl, 5.0; Ca-pantothenate, 20.0; niacin, 40.0; biotin, 0.040; folic acid, 1.80; cyanocobalamin, 0.002; vitamin A acetate (500,000 IU g⁻¹), 2.40; vitamin D₃ (400,000 IU g⁻¹), 0.50; DL-α-tocopheryl acetate, 80.0; and α cellulose, 834.258. ⁸ MP Biomedicals Inc., Solon, OH, USA. ⁹ Stay C^®, (L-ascorbyl-2-polyphosphate 35% Active C), Roche Vitamins Inc., Parsippany, NJ, USA. ¹⁰ Acros Organics B.V.B.A. Thermo Fisher Scientific, Waltham, MA, USA. ¹¹ Alfa Aesar, Ward Hill, MA, USA.

).

Pre-ground ingredients, followed by oil, were added to a floor mixer (Globe Food Equipment Co., Moraine, OH, USA) and mixed for a duration of 15 min. Then, boiling water was blended into the mix until an appropriate consistency was reached. Once mixed, the mash was extruded through a meat grinder with a 3 mm die and then placed in a forced-air drying oven to remove excess moisture, targeting a final moisture of ~10%. Diets were ground and stored at −20 °C. The experimental diets were analyzed at University of Missouri Agricultural Experiment Station Chemical Laboratories (Columbia, MO, USA) for proximate analysis and amino acid profile following AOAC procedures (

Table 2. Amino acid analysis of the experimental diets * with varying levels of methionine offered to Florida pompano for 56 days. University of Missouri Agricultural Experiment Station Chemical Laboratories (Columbia, MO, USA).

Composition (g/100 g, Dry Matter)	0.57% Met	0.66% Met	0.71% Met	0.82% Met	0.86% Met	0.94% Met	1.03% Met	1.28% Met
Alanine	1.81	1.82	1.75	1.79	1.8	1.71	1.83	1.81
Arginine	2.78	2.84	2.65	2.74	2.81	2.67	2.82	2.84
Aspartic Acid	4.00	4.03	3.86	3.94	3.97	3.80	3.97	3.95
Cysteine	0.50	0.51	0.49	0.49	0.50	0.48	0.51	0.50
Glutamic Acid	6.71	6.75	6.48	6.53	6.67	6.22	6.67	6.65
Glycine	2.67	2.63	2.41	2.37	2.57	2.17	2.27	2.03
Histidine	0.93	0.95	0.89	0.92	0.95	0.89	0.93	0.94
Hydroxylysine	0.05	0.06	0.05	0.06	0.06	0.05	0.05	0.06
Hydroxyproline	0.26	0.25	0.26	0.26	0.27	0.24	0.26	0.26
Isoleucine	1.84	1.86	1.76	1.81	1.83	1.76	1.83	1.82
Leucine	2.93	2.96	2.80	2.89	2.93	2.80	2.93	2.91
Lysine	2.49	2.52	2.39	2.45	2.50	2.39	2.48	2.51
Methionine	0.57	0.66	0.71	0.82	0.86	0.94	1.03	1.28
Ornithine	0.06	0.06	0.07	0.07	0.06	0.06	0.06	0.07
Phenylalanine	1.93	1.95	1.85	1.91	1.93	1.85	1.94	1.92
Proline	2.11	2.11	2.04	2.10	2.10	2.03	2.15	2.13
Serine	1.64	1.59	1.57	1.58	1.61	1.54	1.61	1.59
Taurine	0.68	0.70	0.66	0.67	0.70	0.66	0.69	0.73
Threonine	1.46	1.45	1.40	1.43	1.44	1.39	1.45	1.43
Tryptophan	0.54	0.55	0.52	0.53	0.53	0.51	0.52	0.53
Tyrosine	1.15	1.27	1.07	1.17	1.31	1.21	1.30	1.36
Valine	1.95	1.99	1.88	1.93	1.96	1.84	1.99	1.95

* Digestibility coefficient were determined with feeds marked with yttrium. Apparent digestibility of protein was calculated at 78%, while digestibility coefficient of methionine was calculated at 82%.

). Analysis of amino acid profile follows methods 982.30 E (a,b,c) chp. 45.3.05. Crude protein was obtained using method 990.03, 2006. Ash was obtained using method 943.05. Fat was obtained using method 920.31 (A). Fiber was obtained using method 978.10, 2006. Moisture was obtained using 934.01, 2006.

2.2. Experimental Design

Florida pompano juveniles were obtained from Proaquatix LLC (Vero Beach, FL, USA) and were transported to EW Shell Fisheries Center, Auburn, AL, USA. Fish were then transferred indoors and acclimated into a large, fiberglass nursery comprising a reservoir tank and a biofilter. Supplemental aeration was provided using a central line, a regenerative blower, and air diffusers. Fish were fed a commercial diet consisting of 50% protein and 15% lipid to apparent satiation (FF Starter, Ziegler Bros., Inc., Gardners, PA, USA) until the desired size was reached for the experiment. At the beginning of the trial, 15 fish were euthanized using an overdose of buffered MS-222 (Tricaine-S, Western Chemical Inc., Ferndale, WA, USA) and stored at −20 °C for subsequent whole-body proximate analysis. Then, 600 size-sorted fish with a mean initial weight of 4.04 g (±0.05) were moved to a separate system comprising 40 square aquaria (132 L), with 15 fish being stocked per tank. The system comprised a reservoir tank, a bead filter, mechanical and biological filtration, circulation pumps, and supplemental aeration. Diets were randomly assigned to the tanks, allowing for five replicate tanks per diet. Fish were offered experimental feeds based on a percentage (ranging from 5–10%) of the fish’s biomass in a given tank and were fed four times per day (8:00, 10:30, 13:00, 15:30) to apparent satiation. Feed rations were adjusted accordingly based on feeding response, observed growth, and water quality. Fish were weighed on a bi-weekly basis and were dipped in freshwater, then immersed in a chloroquine diphosphate (TCI America, Portland, OR, USA) bath for one minute at each weighing for disease prevention. The feed ration was adjusted after every weighing and as needed based on visual observation of feeding.

Water quality parameters including temperature, salinity, and dissolved oxygen (DO) were measured in the morning and afternoon each day with a water quality multi-parameter meter (YSI ProPlus, Yellow Spring Instruments Co., Yellow Springs, OH, USA), while pH was measured twice a week using a Waterproof Pocket pH Tester with 0.1 Resolution (pHep, Hanna Instruments, Smithfield, RI, USA). Total ammonia nitrogen (TAN) and nitrite were measured twice per week using a YSI 9500 economical photometer (YSI Inc., Yellow Springs, OH, USA). The experimental system was regularly backwashed, and routine partial water exchanges were performed to maintain water quality parameters at desirable conditions for Florida pompano. The system was routinely treated for parasites using copper sulphate pentahydrate at 0.15–0.2 mg/L Cu²⁺ every other week [28,29]. Copper levels present in the water were tested the same day using a Chemetrics copper testing kit (CHEMetrics, AquaPhoenix Scientific, Midland, VA, USA).

The experiment was terminated after fifty-six days. The fish were group weighed and counted to determine final biomass, final mean weight, percent weight gain, thermal growth coefficient (TGC), feed conversion ratio (FCR), apparent net protein retention (ANPR), and survival. Equations used in the calculation of each growth parameter are as follows:

Final Biomass = Average group weight of each treatment

Final Mean Weight = Average individual weight of each fish per treatment

Percent Weight Gain = (Final weight − Initial weight)/Initial weight × 100

TGC = [Final weight^1/3 − Individual weight^1/3/∑Days × °C] × 100

FCR = (Total feed consumed/Weight gain of organism)

ANPR = (Final weight × Final protein content) − (Initial weight × Initial protein content) × 100/protein intake

Survival = (Total number of individuals alive at the end of the trial − Total number of individuals at the start of the trial/100)

2.3. Proximate/Amino Acid Analysis

Upon termination, four fish per tank were haphazardly chosen, euthanized, and stored at −20 °C for whole-body proximate and amino acid analysis. Proximate analysis and amino acid analysis of Florida pompano samples were performed at University of Missouri Agricultural Experiment Station Chemical Laboratories (Columbia, MO, USA). Amino acid analysis and proximate analysis of crude protein, ash determination, crude fat, crude fiber, and moisture were performed in accordance with the Association of Official Analytical Chemists, International (AOAC) Official Methods. The analysis of amino acid profile followed methods 982.30 E (a,b,c) chp. 45.3.05. Crude protein was obtained using method 990.03, 2006. Ash was obtained using method 943.05. Fat was obtained using method 920.31 (A). Moisture was obtained using 934.01, 2006.

2.4. Analytical Modeling

To estimate quantitative methionine requirements, percent weight gain (PWG), thermal growth coefficient (TGC), mean weight (MW), apparent net protein retention (ANPR), and methionine retention (MR) were fitted against dietary methionine levels using four statistical models: one-slope broken-line model (BLM1), two-slope broken-line model (BLM2), broken quadratic model (BQM), and four-parameter saturation kinetic model (SKM-4). BLM1 and BLM2 are fitted by the method of least squares and yield an estimate for dietary requirements, with BLM1 being composed of an ascending or descending slope intersecting with a plateau. Comparatively, BLM2 shows the intersection of two ascending or descending slopes [30,31]. BQM is used to fit curvilinear data and is composed of an ascending or descending quadratic intersecting with a plateau, with values below the requirement fitting the quadratic section and values above the requirement being fit to the plateau. (Robbins, 2006). SKM-4 is a continuous, non-linear model whose requirements are determined at 95% of the maximum observed response [32,33,34]. Mathematically, the equations are expressed as follows:

BLM 1: Y = L + U(R − XLR)

where:

• L: Ordinate of the breakpoint in the curve;
• R: Abscissa of the breakpoint in the curve;
• U: Slope of the line when X is less than R;
• R − X_LR: Equal to zero when X is greater than R;

BLM 2: Y = L + U(R − X_LR) + V(X_GR − R)

where:

• V: Slope of the line when X is greater than R;
• X_GR − R: Equal to zero when X is less than R;

BQM: Y = a + U(X_bp − X)² +V(X − X_bp)

where:

• a: Asymptote of the quadratic ascending portion of the model;
• X_bp: Abscissa of the model’s breaking point. X_bp − X equals 0 when x is greater than X_bp

$$\mathrm{KM}-4: \mathrm{Y}={\displaystyle \frac{(\mathrm{i}\times \mathrm{k}0.5)+(\mathrm{Ymax} \times \mathrm{Xn})}{(\mathrm{k}0.5+\mathrm{Xn})}}$$

where:

• i: Model’s intercept;
• Ymax: Maximum theoretical response;
• k0.5: Level for half of (Ymax + i);
• n: Apparent kinetic order.

2.5. Statistical Analysis

Survival, final biomass, mean weight, overall weight gain, PWG, and FCR were analyzed using a one-way analysis of variance (ANOVA). Significant differences among means were determined using the Student–Neuman–Keuls (SNK) multiple range test. Differences were considered significant at p < 0.05. ANOVA and SNK were performed using SAS for windows (V9.4 SAS Institute, Cary, NC, USA). All variables used in estimating dietary methionine requirements from BKL 1, BKL 2, BQM, and SKM-4 were statistically analyzed using R version 4.2.1 (R Foundation for Statistical Computing, Vienna, Austria) along with the packages “minpack.lm”, “MuMIn”, “boot”, and “nlme.” These packages were used to evaluate model fit and confidence intervals around the estimated requirement. Model selection was primarily based on Akaike Information Criterion (AICc) as well as resulting Akaike model weights. AICc is a commonly used method to measure goodness of fit when comparing and differentiating separate models, with the lower number resulting in a better fit for the model. The Akaike model weights are the probabilities of a model being the model that best fits the data among those being tested [35,36].

3. Results

Water quality parameters (mean ± standard deviation) measured during the growth trial were kept within acceptable ranges for the culture of Florida pompano during the growth trial and are summarized in

Table 3. Results (mean ± standard deviation) of water quality parameters of the experimental system over the course of the 56-day growth trial. Temperature, salinity, and DO were recorded daily in the morning and afternoon. System pH, total ammonia nitrate, and nitrite were all measured twice per week.

Parameters	Mean ± Standard Deviation *
Temperature (°C)	26.94 ± 1.76
Salinity (g/L)	12.45 ± 1.52
DO (mg/L)	6.80 ± 1.54
pH	7.24 ± 0.27
TAN (mg/L)	0.36 ± 0.27
Nitrite (mg/L)	0.39 ± 0.40

* Data are represented as mean values standard deviation.

.

There were no significant differences observed in survival (p = 0.237). However, there were significant differences observed in mean final biomass, individual weight, PWG, TGC, and FCR (p > 0.05). Final biomass of fish reared on the various diets ranged from 136.9 to 272.8 g (0.57 and 1.28% dietary methionine (as-is basis), respectively). The final individual weights of fish reared on the various diets ranged from 9.73 to 18.63 g (0.57% and 1.28% dietary methionine per 100 g of diet (as-is) and PWG ranged from 140.54 to 360.99 g (0.57% and 1.28% dietary methionine (as-is basis)). The mean final biomass, mean weight, and PWG all showed the same level of significant differences among treatments, with fish offered a diet containing 0.57% methionine and 0.66% of dietary methionine (as is basis) being significantly different from those offered the remaining diets. The TGC and FCR also show a similar trend, as the diet containing the two lowest levels of methionine (0.57% and 66%) being significantly different from one another, as well as the remaining diets. Fish offered the diet with the highest level of methionine (1.28% dietary methionine, as is) showed the highest mean final biomass, individual weight, PWG, and TGC, as well as having the lowest FCR. Conversely, fish offered the diet with the lowest level of methionine showed the lowest mean final biomass, mean weight, PWG, and TGC, as well as having the highest FCR (

Table 4. Growth performance of juvenile Florida pompano (average individual weight 4.04 ± 0.05 g) offered diets with varying levels of methionine (0.565–1.280%) over the course of the 56-day growth trial. Growth parameters with different superscripted letters denote significant differences (p < 0.05) between methionine % level offered. Significant differences were determined utilizing a one-way ANOVA followed by Student–Newman–Keuls multiple range test.

Met Level (%)	Biomass (g)	Mean Weight (g)	Weight Gain (%)	TGC 1	FCR 2	ANPR 3 (%)	Survival (%)
0.57% Met	136.9 b	9.73 b	140.5 b	0.03 52 c	4.95 a	8.35 c	94.7
0.66% Met	162.2 b	12.55 b	204.0 b	0.04 74 b	3.61 b	12.33 b	86.7
0.71% Met	223.5 a	16.44 a	315.5 a	0.06 19 a	2.56 c	19.46 a	90.7
0.82% Met	230.7 a	18.32 a	352.0 a	0.06 82 a	2.54 c	19.02 a	84.0
0.86% Met	240.7 a	17.39 a	334.4 a	0.06 61 a	2.60 c	18.48 a	92.0
0.94% Met	230.4 a	16.76 a	316.6 a	0.06 38 a	2.73 c	20.19 a	91.7
1.03% Met	246.5 a	18.08 a	349.6 a	0.06 78 a	2.47 c	19.44 a	90.7
1.28% Met	272.8 a	18.63 a	361.0 a	0.06 96 a	2.43 c	19.76 a	97.3
p-value	0.0001	0.0001	0.0001	0.0001	0.0001	0.0001	0.2372
PSE 4	16.00	0.9968	25.11	0.00344	0.221	1.2025	3.54

¹ Thermal Growth Coefficient. ² Feed Conversion Rate. ³ Apparent Net Protein Retention. ⁴ Pool Standard Error.

).

Whole-body proximate analysis showed no statistical differences in moisture, fat, or ash in any individual fish (

Table 5. Whole-body proximate and amino acid analysis of Florida pompano offered experimental diets containing varying levels of methionine over the course of 56 days. University of Missouri Agricultural Experiment Station Chemical Laboratories (Columbia, MO, USA). Nutrient parameters with different superscripted letters denote significant differences (p < 0.05) between methionine % level offered. Significant differences were determined utilizing a one-way ANOVA followed by Student–Newman–Keuls multiple range test.

Proximate Analysis (g/100 g, As Is)	Initial	0.57% Met	0.66% Met	0.71% Met	0.82% Met	0.86% Met	0.94% Met	1.03% Met	1.28% Met	p-Value	PSE
Crude protein	12.17	15.80 B	17.09 AB	18.07 A	16.98 AB	17.67 A	18.19  A	17.50 A	17.46 A	0.009	0.392
Moisture	80.5	71.24	71.80	70.40	71.80	70.23	71.76	70.92	71.35	0.927	0.888
Crude Fat	4.84	7.60	6.55	7.38	6.19	7.84	6.70	7.23	7.472	0.749	0.684
Ash	2.68	4.47	3.97	3.69	3.93	3.72	3.72	3.59	4.178	0.220	0.238
Composition (g/100 g, as is)
Alanine	0.85	1.09	0.98	1.07	1.10	1.16	0.99	0.98	1.13	0.505	0.079
Arginine	0.82	0.98	0.98	0.96	0.96	1.02	0.84	0.81	0.89	0.719	0.090
Aspartic Acid	1.02	1.37	1.34	1.42	1.33	1.39	1.34	1.31	1.37	0.993	0.093
Cysteine	0.09	0.15	0.14	0.14	0.14	0.14	0.14	0.14	0.15	0.996	0.011
Glutamic Acid	1.53	1.97	1.86	2.00	1.92	1.95	2.00	1.88	1.98	0.975	0.138
Glycine	1.21	1.43	1.29	1.36	1.51	1.64	1.39	1.20	1.49	0.356	0.469
Histidine	0.24	0.31	0.31	0.33	0.28	0.30	0.32	0.29	0.32	0.974	0.026
Hydroxylysine	0.07	0.04	0.04	0.04	0.04	0.05	0.05	0.03	0.04	0.370	0.007
Hydroxyproline	0.33	0.40	0.34	0.36	0.43	0.51	0.39	0.30	0.41	0.764	0.069
Isoleucine	0.48	0.66	0.68	0.71	0.64	0.71	0.66	0.74	0.71	0.522	0.034
Leucine	0.77	1.06	1.03	1.13	1.01	1.11	1.05	1.13	1.09	0.848	0.061
Lysine	0.86	1.19	1.11	1.26	1.19	1.21	1.27	1.28	1.24	0.363	0.055
Methionine	0.33	0.41	0.40	0.42	0.40	0.42	0.40	0.39	0.41	0.992	0.028
Ornithine	0.02	0.07	0.07	0.10	0.08	0.09	0.18	0.17	0.15	0.084	0.035
Phenylalanine	0.48	0.62	0.64	0.64	0.63	0.66	0.60	0.65	0.63	0.944	0.031
Proline	0.73	0.89	0.81	0.86	0.92	1.01	0.85	0.75	0.90	0.415	0.088
Serine	0.41	0.56	0.55	0.57	0.55	0.57	0.52	0.51	0.53	0.985	0.045
Taurine	0.10	0.38 B	0.42 AB	0.43 AB	0.41 AB	0.42 AB	0.45 AB	0.43 AB	0.47 A	0.030	0.017
Threonine	0.48	0.66	0.63	0.67	0.64	0.67	0.66	0.66	0.69	0.975	0.034
Tryptophan	0.10	0.16	0.17	0.19	0.19	0.18	0.19	0.18	0.19	0.395	0.008
Tyrosine	0.31	0.43	0.45	0.44	0.45	0.44	0.40	0.39	0.39	0.737	0.031
Valine	0.56	0.75	0.73	0.78	0.74	0.78	0.77	0.80	0.78	0.738	0.031

). There was a significant statistical difference in the crude protein levels (p = 0.008) of fish offered diets with 0.57% dietary methionine (15.80%) (as is) and fish offered diets with 1.28% dietary methionine (17.46%) (as-is). The amino acid composition of whole-body fish showed no statistically significant differences in the presence of amino acids in individuals, with the exception of taurine (p = 0.0301). Different levels of taurine were observed in the bodies of fish offered diets with 1.28% dietary methionine (0.47%) (as is) and fish offered 0.57% dietary methionine (0.38%) (as-is).

Model fitting for PWG, MW, TGC, ANPR, and methionine retention (MR) are presented in

Table 6. Regression model selection used in calculating the dietary methionine requirements in Florida pompano reared over the course of 56 days. Model selection was determined using R² values, AICc values, and Akaike weights obtained in R Studio.

Parameter	Model	R2	AICc	AIC Weights	Requirement Dry Weight 95% CI
TGC 1	BLM1	0.71	−271.45	0.432	0.74 (0.696–0.777)
	BLM2	0.71	−269.35	0.152	0.73 (0.681–0.775)
	BQM	0.59	−267.43	0.058	0.83 (0.714–0.943)
	SKM-4	0.66	−271.07	0.358	0.71 (0.682–0.978)
PWG 2	BLM1	0.65	−441.15	0.384	0.75 (0.690–0.797)
	BLM2	0.65	−440.96	0.134	0.74 (0.677–0.794)
	BQM	0.51	−444.85	0.060	0.82 (0.678–0.971)
	SKM-4	0.63	−443.26	0.422	0.72 (0.686–1.367)
MW 3	BLM1	0.64	−184.22	0.467	0.75 (0.693–0.804)
	BLM2	0.65	186.28	0.167	0.74 (0.675–0.801)
	BQM	0.40	187.49	0.091	0.84 (0.693–0.993)
	SKM-4	0.53	185.28	0.275	0.72 (0.689–1.238)
ANPR 4	BLM1	0.70	199.41	0.240	0.72 (0.690–0.753)
	BLM2	0.49	201.68	0.077	0.71 (0.673–0.756)
	BQM	0.46	205.03	0.014	0.81 (0.704–0.913)
	SKM-4	0.66	197.36	0.668	0.68 (0.680–0.716)
MR 5	BQM	0.10	246.94	0.96	0.76 (0.686–0.829)
	SKM-4	0.16	253.29	0.04	0.89

¹ Thermal Growth Coefficient. ² Percent Weight Gain. ³ Mean Weight. ⁴ Apparent Net Protein Retention. ⁵ Methionine Retention.

. BLM1 was shown to be a good fit for MW (AICc = 184.223 and Akaike = 0.467) and TGC (AICc = −217.441 and Akaike = 0.432). The results of TGC being fitted with BLM1 show a methionine requirement of 0.74% with a range between 0.70 and 0.78% (Figure 1). AICc and Akaike weights demonstrate a goodness of fit for the model SKM-4 for PWG (AICc = 443.257 and Akaike = 0.422), as well as apparent net protein retention (AICc = 197.362 and Akaike = 0.688). The results of ANPR fitted to the SKM-4 model show a methionine requirement of 0.68% with a range of 0.68 and 0.72% (Figure 2). BQM was an overwhelmingly good fit for methionine retention (AICc = 246.94 and Akaike = 0.96) and predicted a methionine requirement of 0.76% with a range of 0.68 to 0.82% (Figure 3). R was unable to calculate AICc and Akaike weights for BLM 1 and BLM2 for methionine retention, as well as the requirements for BLM1, BLM2, and SKM-4. Estimates for methionine requirements were consistent across all models, with the exception of the SKM-4 calculation for methionine retention, which placed the value at 89%. Estimates calculated by BQM also placed methionine requirements higher than the other models, between 0.81 and 0.85% (Table 6).

Retention of EAAs in whole-body fish offered experimental diets was shown to be significantly different (p < 0.05) in all EAAs apart from arginine (p = 0.0705). Fish offered diets with the lowest levels of methionine (0.57% and 0.66% respectively) were shown to have significantly lower retention levels for leucine, lysine, phenylalanine, threonine, tryptophan, and valine when compared to fish offered the other diets. Similarly, retention of histidine was shown to be significantly lower between fish offered the diet with the lowest level of methionine (0.57%) and fish offered the highest level of methionine (1.28%). Isoleucine retention was significantly lower in individual fish offered the diet with 0.57% methionine when compared to fish offered diets with methionine levels ranging from 0.71% and 1.28%. Differences in methionine retention were observed in fish offered 0.82% methionine and fish offered 1.28% methionine, with individuals offered 0.82% methionine having the highest retention levels and individuals offered 1.28% methionine having the lowest retention levels. Taurine retention followed a somewhat similar trend to the other EAAs, with the diets containing the lowest levels of methionine (0.57% and 0.66%) being significantly different than the diets containing methionine levels between 0.71% and 1.28%. However, diets one (0.57% methionine) and two (0.66% methionine) were also shown to have significantly different levels of taurine retention between the two of them as well (

Table 7. Retention of essential amino acids (EAAs) in whole-body Florida pompano offered experimental diets with graded levels of dietary methionine. Parameters within a row with different superscripted letters denote significant differences (p < 0.05) between methionine % level offered. Significant differences were determined utilizing a one-way ANOVA followed by Student–Newman–Keuls multiple range test.

Composition (g/100 g, Dry Weight)	0.57% Met	0.66% Met	0.71% Met	0.82% Met	0.86% Met	0.94% Met	1.03% Met	1.28% Met	p-Value	PSE
Arginine	7.06	9.42	13.90	14.84	14.88	11.72	11.62	12.82	0.0705	2.0936
Histidine	7.10 b	8.83 ab	15.31 a	13.19 ab	13.27 ab	15.14 a	13.64 ab	14.10 a	0.0076	1.8047
Isoleucine	8.28 c	11.48 bc	17.68 a	15.91 ab	16.66 a	15.74 ab	18.62 a	17.68 a	0.0001	1.4719
Leucine	8.27 b	10.40 b	17.54 a	15.66 a	16.26 a	15.86 a	17.68 a	16.82 a	0.0001	1.5903
Lysine	10.89 b	12.63 b	23.51 a	22.12 a	20.67 a	23.00 a	23.76 a	22.41 a	0.0001	1.7768
Methionine	15.51 ab	15.96 ab	24.61 a	21.08 ab	20.04 ab	17.01 ab	16.40 ab	13.93 b	0.0172	2.3106
Phenylalanine	6.87 b	9.51 b	14.59 a	14.64 a	14.35 a	13.26 a	14.85 a	14.37 a	0.0001	1.2345
Taurine	19.16 c	24.60 b	35.27 a	33.63 a	30.54 a	35.50 a	34.31 a	34.89 a	0.0001	2.3190
Threonine	10.28 b	12.42 b	20.23 a	20.27 a	20.05 a	20.27 a	20.48 a	20.89 a	0.0001	1.8268
Tryptophan	8.00 b	10.64 b	17.08 a	17.86 a	15.92 a	16.89 a	17.21 a	17.67 a	0.0001	1.5030
Valine	8.57 b	10.83 b	17.75 a	17.32 a	17.00 a	17.81 a	18.48 a	17.96 a	0.0001	1.4205

).

4. Discussion

Methionine is often considered one of the first limiting amino acids in feed formulations composed of plant-based proteins. In this experiment, a growth trial was performed over the course of 56 days utilizing an experimental diet composed of red lentil meal. Red lentil meal was selected as a primary protein source due to its high protein content and favorable amino acid profile, while having very low levels of methionine. Previous, unpublished work within the lab placed methionine requirements for Florida pompano at between 0.70 and 0.80% of diet (as-is basis). Therefore, eight experimental diets were formulated and supplemented with graded levels of crystalline methionine, resulting in a range between diets of 0.57 to 1.28% of the diet (as-is basis).

It should be noted that alongside the unpublished reports from our lab, there are also two previous reports of trials evaluating methionine requirements in Florida pompano. One is the thesis work of Belfranin (2016), who utilized practical diets supplemented with methionine ranging from 0.50 and 1.50% of diet in 200 g Florida pompano. No adverse effects in growth across all dietary treatments were observed, most likely due to limited tissue turnover (<150% weight gain). The other report is found in an abstract from Louisiana State University (LSU) [27], which utilized six experimental diets supplemented with graded levels of methionine at 0.44–2.04% of the diet. Weight gain data indicated the optimum dietary methionine requirement of Florida pompano to be 1.17% (2.54% of CP) when fitted against a broken-line regression (R2 = 0.72) and 1.60% (3.48% of CP) based on second-order polynomial regression (R2 = 0.71).

Within this study, Florida pompano exhibited the least amount of growth as well as the highest FCR when offered diets containing the lowest levels of methionine, between 0.57 and 0.66% of the diet, respectively (as-is basis). It is important to note that in most growth studies involving Florida pompano, FCR is usually higher than in other cultured species due to the pompanos feeding habits as well as the way in which they feed. While this may cause FCR values to be overestimated, they remain relevant to EAA retention. Reduction in growth parameters can be attributed to diets lacking the required level of dietary methionine. Because methionine is a main donor of methyl, it is closely linked to many important metabolic functions [37]. A reduced metabolism would result in decreased growth, as seen in fish offered these diets, and would explain why growth factors improved upon meeting the dietary methionine requirement. This can be seen in a study conducted by Mai et al. (2006), where yellow croakers (Pseudosciaena crocea) were offered diets containing varying levels of methionine. According to Mai et al., the level of methionine present in the experimental diets contributed to or impeded specific growth rate, feed conversion efficiency, and protein efficiency ratio, with all values increasing with increasing levels of methionine, before eventually decreasing upon reaching the assumed requirement threshold [38]. A similar study conducted by Michelato, Furuyu, & Gatlin (2018) showed similar effects in Nile tilapia, with fish offered lower than presumed amounts of methionine had impeded growth compared to individuals offered diets with desired amounts of methionine or diets lacking in methionine that were also supplemented with taurine [39]. This study also further demonstrates a relation between the sulfur amino acids and taurine in the diets of cultured finfish.

Taurine is not considered an essential amino acid in the diets of many fish species due to its ability to be biosynthesized within the body [40]. However, biosynthesis of taurine varies between species [40,41]. In some cases, marine teleosts demonstrate a poorer capacity for taurine synthesis than those of freshwater species such as rainbow trout [40,41]. This has led taurine to be labeled a conditionally essential amino acid in fish nutrition [42]. Taurine itself is an amino sulfonic acid, containing an acidic sulfonate group, a basic amino group, and two carbon atoms [43], and is responsible for a range of important functions [42]. In fish, taurine is typically synthesized from cysteine through the cysteine sulfinic acid decarboxylation pathway, with cysteine being a product of methionine catabolism [44,45]. Some of the experimental diets in this study were, by design, deficient in methionine. With methionine being in such short supply in some of the experimental diets, this could lead to statistically significant differences in taurine observed across whole-bodied proximate analysis as well as the decrease in growth.

Results of the whole-bodied proximate analysis of fish also showed a statistically significant difference in crude protein levels of fish offered the basal diet as opposed to fish offered diets with a dietary methionine range of 0.70% to 1.28% (as-is basis). Methionine is an essential amino acid in the biosynthesis of proteins in eukaryotes, acting as the initiating amino acid for protein synthesis [46,47]. However, the analyzed fish showed no statistical differences in the level of methionine present in the protein. The only AA which shifted in the whole-body analysis was taurine, which tended to increase. Taurine is one of the few amino acids that do not play a role in the synthesis of proteins; it instead may increase the levels of protein found in the bodies of mitochondria [48]. It also affects protein folding through a reduction of oxidative stress [49] and, in mice, was shown to be important in protein stabilization [50]. So, while taurine may not assist in the biosynthesis of protein, it is shown to interact with and manipulate protein in many ways. We theorize that this could potentially explain why fish with significant differences in crude protein within their bodies also exhibited significant differences in taurine.

Statistical modeling has been used for several decades to determine nutritional requirements in several important species [51,52,53,54,55]. Model selection for the five growth parameters (PWG, MW, TGC, ANPR, MR) estimated a minimum dietary methionine requirement for Florida pompano of 0.68% (as-is basis). Previous data from the Auburn University aquatic animal nutrition lab confirm the obtained results (data not published). This would place the dietary methionine requirements of Florida pompano much lower than those of similar species of marine finfish, such as the silver pompano (1.16–1.18%) [56], as well as the golden pompano (1.06–1.27%) [57]. Wang et al. described most species possessing a methionine requirement between 0.49 and 2.50% of the diet, or 1.49 and 4.70% of dietary protein. Thus, the methionine requirements estimated from this study are within the expected range for a commercially important finfish species [18].

TGC is a commonly used calculation in fish growth trials and was fitted against dietary methionine to determine appropriate methionine levels needed for growth. In Figure 1, the BLM1 model used to fit TGC (R2 = 0.71) places the requirements at 0.74% with a 95% confidence interval (CI) between 0.70 and 0.78%. These results place the predicted dietary methionine requirement within the range of all other statistical models (with the exception of ANPR). Likewise, Figure 2 shows that the data for ANPR was best fit by the SKM-4 model (R2 = 0.60) and yielding a requirement prediction of 0.68% (95% CI of 0.68–0.72). While MR requirement estimation was placed at 0.76% for methionine, its shape differed markedly from the other four growth parameters. All data for growth parameters were fitted with either a BLM1 or SKM-4 statistical model, composed of an ascending slope intersecting with or leveling off into a plateau. This is due to the requirement usually being met at or close to the intersection of the two lines. In the case of the data for MR, these data do not form a plateau due to methionine being found in excess past its requirement. As a result, the data take on a quadratic shape, with the requirement being found near the top of the data. Figure 3 shows the data peaks just before 0.80% dietary methionine, demonstrating that the requirement may be found near this point, which is further supported by the results of the other growth parameters. The data for methionine retention, as well as the remaining EAAs, are summarized in Table 7.

In general, retention levels of EAAs appear to follow a similar trend, where retention values increased up to the requirement for methionine, then plateaued. This is, in part, due to methionine having the ability to reduce the oxidation rates of other EAAs [17] as well as the imbalance of EAAs. Unlike the other AAs, which were not supplemented, higher levels of methionine resulted in a decrease in retention values. This is due to methionine being in excess within the diet and no longer being utilized efficiently within the body. Excess methionine impedes the transamination pathway that leads to its metabolism [58]. The diet that supported the highest level of methionine retention in the fish contained 0.82% methionine. While this is slightly higher than the projected requirement within the present work, it is still within the projected ranges shown by the SKM-4 model for PWG (0.686–1.367) as well as the BQM model for methionine retention (0.686–0.829). Using methionine as a benchmark, we can see that diets supplemented with 0.71% methionine had a retention efficiency of 24.61%. In the same diet, lysine (23.51%) and taurine (35.27%) also had high retention efficiencies. Typically, a high retention value indicates that they are close to the requirement; hence, these are the most limiting AAs. The next numerically high retention value was that of threonine, which would indicate that it is likely the next most limiting AA within this dietary matrix and a requirement should be pursued.

5. Conclusions

Based on the results obtained from the statistical models of growth parameter data collected during this study, we project the minimum methionine requirements for juvenile Florida pompano to be around 0.68% of the diet or 1.70 g/100 g protein. TGC, when fitted to BLM.1, specifically places the requirement at 0.74%. This is further supported by the MW and MR, which place the methionine requirement at 0.75% and 0.76%, respectively. Individuals offered diets including this amount of dietary methionine or more grew to a greater size than those offered diets including less. Based on these findings, we recommend offering juvenile Florida pompano diets containing methionine at this level to ensure optimal growth. Additionally, based on AA retention values, we would recommend that threonine may be the fourth limiting amino acid and warrants research.