Utilization of Fish Meal and Fish Oil from Smoked Salmon By-Products in Juvenile Striped Bass (Morone saxatilis) Feeds: Growth Performance, Nutritional Composition, and Shelf-Life Assessment of Upcycled Ingredients

Abstract

Fish meal (FM) and fish oil (FO) are vital components commonly used in feed formulations. However, their supply, which generally comes from capture fisheries, is being exhausted, necessitating the exploration of sustainable alternatives. In a two-part study, the first part evaluated the FM and FO derived from smoked salmon by-product (SSBP) over a 12-week accelerated shelf-life test, comparing their lipid oxidation, amino acid, and fatty acid profiles to those of commercial whitefish meal and oil. In the second part, the SSBP FM and FO were then included in three experimental feeds at 25%, 50%, and 100% inclusion levels. These feeds were tested on juvenile striped bass (Morone saxatilis) cultured in a recirculating aquaculture system (RAS). The results indicated that the quality of SSBP FM and FO was lower than the commercial product (less amino acids (23.98% vs. 60.30%) and omega-3 fatty acids (9.46% vs. 26.6%), respectively). SSBP FO exhibited high initial peroxide value (21.00 ± 0.00 meq/kg oil), with gradually increasing total oxidation value and p-Anisidine value during storage. Regarding the feeding trial, all fish showed signs of Mycobacterium marinum infection after one month. While there was no significant difference in feed palatability (p > 0.8559), the feed conversion ratio was less efficient for the 100% SSBP feed (1.44 ± 0.14) compared to commercial feed (1.36 ± 0.13), but these differences were not statistically significant. This study suggests that SSBP FM and FO can be used as supplements at lower levels (25% and 50%) without negatively affecting growth, feed efficiency, or survival. Our findings may be useful for enabling beneficial collaborations between smoked salmon processors, feed manufacturers, and striped bass farmers, therefore contributing to sustainability in aquaculture practices.

1. Introduction

The diet of animals plays a crucial role in how they develop and grow. As such, when looking at the factors influencing the growth of domestic livestock, including cultured fish, it is important to assess the quality of the feed provided, its nutritional composition, and subsequent feed palatability to ensure that they are eating and growing as efficiently as possible. In the case of most aquacultured fishes, the driving factor in feed quality is centered on the nutritional profile of both the protein and energy sources, that is, the composition of amino acids and fatty acids, respectively. The requirement of specific dietary amino acids varies among many fishes [1]. To meet these needs, many aquaculture feeds incorporate specific amounts of fish meal (FM) as a source of high-quality protein in their formulations. Sulfur-containing amino acids, such as methionine and cysteine, are essential to fish growth and are often considered a limiting source in feeds. The presence of certain other amino acids has also been demonstrated to improve palatability depending on their concentration; for example, alanine and leucine have been demonstrated to stimulate feeding responses in rainbow trout (Oncorhynchus mykiss) [2]. It should be noted that this response to these particular amino acids may be a specialized sensory response by rainbow trout, with other fish specializing in the detection of different amino acids, thereby encouraging niche specificity within teleosts [2]. Plant-based protein feed ingredients are economical when compared to FM. However, care must be taken when choosing the cheaper plant-based materials as they can exhibit reduced palatability and resultant feed intake in fishes and generally should only be used on a case-by-case basis after confirming appropriate baseline inclusion levels for the targeted species [3]. In addition to reduced palatability, plant-based proteins generally lack in essential amino acids such as the well-known limitations of lysine in wheat [4]. Soy proteins, which are a common plant-based protein used in feed, have the potential to limit fish growth as well due to limited sulfur-amino acids [5] and other additional anti-nutritional factors.

Fish oil (FO), on the other hand, has not demonstrated any differences in growth when compared to plant-based alternatives [6,7]. A variety of plant-based oils, such as soybean oil and sunflower oil, have resulted in similar growth and feed efficiency when compared to FO [8]. However, unlike freshwater fish, marine species have lower ability to convert alpha-linolenic acid (ALA, 18:3n3) to eicosapentaenoic acid (EPA, 10:5n3) and docosahexaenoic acid (DHA, 22:6n3) [9,10]. Marine fish, such as turbot (Scophthalmus maximus), red seabream (Chrysophrys major), Asian seabass (Lates calcarifer), etc., cannot effectively elongate or desaturate 18-carbon fatty acids and thus require the presence of highly unsaturated fatty acids (HUFA), especially DHA and EPA in diets to maintain normal metabolism and growth [11]. In contrast, salmonids can convert 18-carbon fatty acids to longer-chain, more unsaturated fatty acids and, thus, 18:3(n−3) and 18:2(n−6) or both are important dietarily [12]. Therefore, FO has a larger role to play as compared to plant-based oils in the overall lipid nutrition of marine and anadromous fish species. In looking at the fatty acid profile of FO, the proportion of omega-3 highly unsaturated fatty acids is much higher compared to that of plant-based oils, where omega-6 and omega-9 fatty acids are more prevalent [13]. This warrants consideration during feed formulation because lipid uptake has the biggest impact on body composition of the animal eating it [14], meaning fish that eat higher proportions of plant-based oils compared to FO have a reduced omega-3 concentration within their tissue. This has the potential to be problematic from a marketing standpoint, as consumers of fish within the USA often do so for the consumption of “healthy fats” found in fish, in regard to omega-3 and omega-6s [15].

Another important characteristic of an ingredient in feed formulation is its ability to have an extended shelf life. In the case of accelerated shelf-life studies, a given product is stored in a range of conditions such as humidity, temperature, and pH deemed to place increased stress on that product. From data on the observed degradation rate and its relationship to which the conditions accelerated it [16], predictions can be made for recommended storage conditions. For example, if a product’s degradation was accelerated by increased temperatures, then it may be recommended to refrigerate said product. FM and FO are susceptible to spoilage if not kept in cold and non-humid conditions. They have relatively poor oxidative stability when compared to plant-based alternatives. This is due to the relatively higher concentration of polyunsaturated fatty acids (PUFAs), which are prone to oxidation. PUFAs contain bis-allylic carbons that require a low activation energy for the creation of free radicals, resulting in the degradation of the fatty acids and, ultimately, the oil [16]. During oxidation, cleavage of PUFAs results in aldehydes and lipid peroxides which are indicators of spoilage. p-Anisidine and peroxide value (PV) are the standard metrics for detecting oxidation during shelf-life assessments [16]. FM and FO are known to have detectable odors at low volatile concentrations (0.01 ppb) during oxidation. The oxidation of oil during production, storage, and use of FO due to the presence of unsaturated fatty acids produces harmful substances like ketones, aldehydes, and hydroperoxide [17]. Several studies have shown the negative impact of oxidized fish oil like growth inhibition [18], nutritional composition change [19], lipid metabolism disorder [20], oxidative stress [21] etc. on aquatic animals. Ultimately, treatment of FM and FO with antioxidants and how they are stored, are the driving factors in prolonging the shelf life of these products. Salmon FM protein degradation can be expected around 12 months at 30 °C [22].

Striped bass was chosen for this study because it is a popular fish along the Eastern and Western Coasts of the United States. Nearly 47% of the total aquaculture sales in the U.S. is accounted by foodfish category and hybrid striped bass with 5% of market share within the foodfish category [23] in the U.S. is produced more than striped bass. Further, there is demand for marine finfish with a desired size of 1.3 to 2.2 kg (3 to 5 lbs.) along the mid-Atlantic region. This demand cannot be met adequately with hybrid striped bass, as their growth rate and feed efficiency significantly decline once they reach approximately 0.6 kg (1.5 lbs.). In contrast, striped bass exhibit superior growth performance and are therefore better positioned to meet the market demand for larger-sized fish [24]. The striped bass and its hybrids feed at the rate of 3% of the body weight per day [25] and the feed conversion ratio (FCR) for hybrid striped bass ranges between 1.5 to 1.7, which is higher than that of coldwater fish, which have an FCR often of about 1.0 [26]. In this study, feeds manufactured with smoked salmon by-product (SSBP) FM and FO were fed to aquacultured striped bass (Morone saxatilis) to evaluate their growth performance when compared to fish fed a commercial feed. Striped bass are opportunistic carnivores that have a wide range of available prey options in the wild due to their anadromous life history, which makes them generalist feeders [27]. This generalist appetite, their ability to withstand a range of salinities, and their appeal to the human palette makes them an excellent fish for whitefish aquaculture [28]. Many feeding trials have been performed on aquacultured striped bass to explore which formulations may be optimal for cultivation of these fish. In one such study with juvenile striped bass, high oleic soybean meal was incorporated as a plant-based protein and lipid alternative. It was found that at 50% replacement of FM, the feed conversion ratio (FCR) was efficient at a range of 1.04 to 1.16 across treatment groups [29]. Similarly, ref. [30] found that supplementation of FM-based fish feed with poultry or plant-based alternatives could reduce cost of manufacturing while maintaining growth performance of hybrid striped bass; however, fish feed supplemented with poultry by-product outperformed plant-based alternatives in terms of percent weight gain, with fish gaining 200–300% more weight [30]. This indicates that striped bass may have a harder time converting plant-based material into body weight over time. Ultimately, cheaper meat-based meals lend themselves better to uptake by the fish due to their life history as carnivores [31].

The study therefore tested the following hypotheses: (1) the inclusion of SSBP FM and FO at varying levels (25%, 50%, and 100%) does not negatively affect the growth performance of juvenile striped bass compared to a commercial feed, (2) FM and FO produced from SSBP have a shelf-life comparable to or better than commercial FM and FO under accelerated shelf-life conditions, and (3) feeds formulated with increasing levels of SSBP FM and FO are equally or more palatable to juvenile striped bass as compared to feeds made with commercial FM and FO. Thus, the objectives of this study were as follows: (1) to evaluate the growth performance of juvenile striped bass fed feed containing SSBP FM and FO, (2) to determine the shelf life of SSBP FM and FO, and (3) to assess palatability of three experimental feeds containing SSBP FM and FO for juvenile striped bass.

2. Materials and Methods

2.1. Control and Experimental Feed Manufacturing and Formulation

Four floating extruded aquaculture feeds were manufactured by Zeigler Bros Feed Company, Inc. (Gardners, PA, USA). All diets were formulated to be isonitrogenous (45%) and isolipidic (11%), while adhering to standard nutritional requirements of juvenile striped bass [32,33]. Formulations were composed using Concept 5 version 10.01 software and is presented in

Table 1. Ingredient composition of control and experimental feeds.

Ingredients	Commercial (%)	25% SSBP (%)	50% SSBP (%)	100% SSBP (%)
Wheat Flour Bagged	23.71	23.71	23.71	23.71
Poultry By-Product Meal: 67	19.65	19.65	19.65	19.65
Soybean Meal: 47.5	18.00	18.00	18.00	18.00
Pacific Whiting Fish Meal 68%	15.00	11.25	7.50	0.00
Corn Gluten 60%	12.29	12.29	12.29	12.29
Menhaden Gold Oil Mixer	4.26	3.20	2.13	0.00
Feather Meal	3.00	3.00	3.00	3.00
Menhaden Gold Oil Topdress	2.00	1.50	1.00	0.00
Dicalcium phosphate	1.00	1.00	1.00	1.00
L Lysine 98.5%	0.25	0.25	0.25	0.25
Choline CL 60%	0.20	0.20	0.20	0.20
Premix Aqua-Vit	0.20	0.20	0.20	0.20
Premix Aqua-Min Fish	0.20	0.20	0.20	0.20
AMONEX Aqua Dry: Mold Inhibitor	0.20	0.20	0.20	0.20
Vitamin C Monophosphate 35%	0.04	0.04	0.04	0.04
SSBP Fish Meal	0.00	3.75	7.50	15.00
SSBP Fish Oil Mixer	0.00	1.07	2.13	4.26
SSBP Fish Oil Topdress	0.00	0.50	1.00	2.00

SSBP = smoked salmon by-product; Premix Aqua-Vit and Premix Aqua-Min Fish are the proprietary vitamin and mineral mix of the manufacturer, Zeigler Bros Inc., Oil mixer = oil added to the feed at the mixer before extrusion; Oil topdress = oil added onto pellets after extrusion.

and the proximate composition of the feeds is provided in

Table 2. Proximate composition of control (commercial) and experimental fish feeds.

Ingredients	Commercial	25% SSBP	50% SSBP	100% SSBP
Crude protein (%)	45.06	45.31	45.56	46.07
Crude fat (%)	11.12	11.15	11.19	11.25
Crude fiber (%)	1.56	1.60	1.64	1.71
Ash (%)	8.08	7.95	7.82	7.55
Moisture (%)	8.00	8.00	8.00	8.00

SSBP = smoked salmon by-product.

. In total, there was one control feed and three experimental feeds. The control feed was Zeigler Bros Inc.’s commercial aquaculture feed.

The experimental feeds had three different inclusion ratios of the SSBP FM and FO. Both of these byproducts were manufactured at North Carolina State University’s Center for Marine Sciences and Technology (CMAST) Seafood Lab [34]. The ratios were 25% SSBP FM and FO to 75% commercial FM and FO, 50% SSBP FM and FO to 50% commercial FM and FO, and 100% SSBP FM and FO. For the feed manufacturing, all ingredients were mixed in a vertical mixer and blended and then ground by a small hammer mill to a particle of <500 µm. Feeds were extruded through a Wenger TX-57 twin-screw extruder (Sabetha, KS, USA), with a screw diameter of 57 mm and a L/D ratio of 24:1 at an installed power of 250/188 hp/kW. The extruder die diameter was 2.3 mm (to achieve a feed size of 3.0 mm), with material being fed at an average feed rate of 146.38 kg/hr. Feeds were dried using a continuous dryer and then sieved with a mesh that has a size of 1680 µm, removing any fine material. From here, the feed was then top-dressed in a 45.35 kg (100 lb) stainless steel mixer. The feed then was coated in oil in a 45.35 kg (100 lb) stainless steel paddle mixer with the oil being added manually via spraying after measuring the oil quantity. The final feed products were then packaged in 18.14 kg (40 lb) bags and shipped to North Carolina State University’s Marine Aquaculture Research Center (MARC, Smyrna, NC, USA).

2.2. Accelerated Shelf-Life Study

FM (125 g) and FO (125 g) that were manufactured at CMAST in accordance with the optimized parameters presented in [34] were shipped to Eurofins Scientific Nutrition Analysis Center (Des Moines, IA, USA) for a 12-week accelerated shelf-life study, corresponding to 12 months of real life storage. The samples were held at 40 °C and 75% RH during storage. The samples were measured in triplicate for p-anisidine value (p-AV), free fatty acids (FFA) (%), and peroxide value (PV) (meq/kg oil) every two weeks with the exception to the second week of the shelf-life study. These tests were conducted under the protocols AOCS Cd 18-19 for p-Anisidine, AOAC 940.28 and AOCS Ca 5a-40 for free fatty acids, and AOCS Cd 8-53 for peroxide values. Total oxidation value (TOTOX) was calculated by the formula from the Codex Alimentarius Commission [35]:

$$\mathrm{TOTOX}=2\times \mathrm{Peroxide} \mathrm{Value}+\mathrm{Anisidine} \mathrm{Value}$$

(1)

2.3. Fish Husbandry

All methods and procedures regarding live animal use were approved by the North Carolina State University Institutional Animal Care and Use Committee (IACUC # 23-353). A total of 600 juvenile domesticated striped bass (Morone saxatilis) were delivered from the North Carolina State University Pamlico Aquaculture Field Laboratory (PAFL, Aurora, NC, USA) to MARC and were stocked 2 weeks prior to the start of study. The average initial body weight of the juvenile striped bass was ~30 g and the average initial length was 140 mm. These fish were randomly selected to stock in 12 circular polyethylene tanks. These 600 fish were randomly divided among the 4 dietary treatments with 3 replicates per treatment in 12 tanks. Fish were placed, 50 fish per tank at a water depth of 763 mm (30 inches). The 757.08 L (300 gallon) tanks were 1651 mm (65 inches) in diameter. All 12 tanks operated as part of a re-circulation type aquaculture system (RAS), where water was filtered using a shared particulate filter, biofilter and UV sterilizer (AquaDyne, Hartwell, GA, USA). Fish were fasted until the start of the palatability study 2 days after stocking, at which point all animals were fed to satiation twice a day. This was due to preliminary observation of nonexperimental fish readily accepting all of the manufactured feeds. Fish were fed two times daily to satiation, with feed buckets initially being filled to 3–5% body weight. Feeding occurred at 8:00 am and 4:00 pm for 150 days. Environmental parameters including dissolved oxygen (DO), temperature, salinity, pH, total ammonia nitrogen (TAN), and alkalinity were maintained to meet the requirements for juvenile striped bass [36]. These requirements were as follows: DO, 7.68 ± 0.44 mg/L; Temperature, 21.31 ± 0.64 °C; Salinity, 9.27 ± 4.87 ppt) and weekly (pH, 7.11 ± 0.14; TAN, 0.28 ± 0.05 mg/L; Alkalinity 289.07 ± 28.96 ppm). A natural photoperiod (12 h light and 12 h dark, with lights set on a timer from 7:15 am to 7:15 pm) was maintained, and any fish mortality was noted and removed.

2.4. Pellet Palatability and Floatability

Fish were fed twice a day at 8:00 am and 4:00 pm with diets according to their randomized designations. Feed (100 g) was placed in each of 12 buckets, one assigned to each tank, where animals were incrementally fed to apparent satiation (this is a period which occurs about 30–40 min after the feed is placed in the tanks. Fishes are continuously observed for feed response and as the feed response slows, the feed addition is decreased carefully to ensure that there is no leftover feed at the end of feeding that needs to be removed from the tanks). Total feed intake in grams was recorded for each feeding session. For testing pellet floatability, 4 buckets of 18.92 L (five gallons) each were filled halfway with water. Pellets (100) of each feed type were counted and held in a small bucket. Each of these small buckets with the 100 pellets was poured from the rim of the 18.92 L (five gallons) bucket into the water. As soon as they hit the water, a 600 s (10 min) timer started. Any pellets that sank were counted. This was repeated in triplicate.

2.5. Feeding Trial Sampling Procedure

Ten fish were sampled from each of the 12 tanks every 28 days for 150 days (6 sampling periods). During sampling, fish were anesthetized with tricaine methanesulfonate (Syndel Laboratories, Ferndale, WA, USA) added at a rate of at least 500 mg per 3.78 L (1 gallon) in the water to a bucket containing 75.70 L (20 gallons) of water. The amount of tricaine methanesulfonate added varies according to environmental and biological factors [37]. The total body weights (g) and lengths (mm) were noted. This information was also used, and coupled with daily recordings of feed intake, to calculate weight gain (WG), percent weight gain (WG%), specific growth rate (SGR), feed conversion ratio (FCR), and condition factor (K). Personal protective equipment was worn during handling (nitrile gloves of ANSI/ISEA 105 standard) due to an outbreak of Mycobacterium marinum during the experiment. At the end of the feeding trial, fish were removed from the feed for 24 h and batch weighed for the determination of final body weight, and other growth performance indicators were calculated. Additionally, 10 fish per tank were randomly sampled for fillet and whole-body composition. These fish were placed in a cold-water bath for humane euthanizing. Afterwards, fish were frozen in chilled water, whole, submerged for 20 min. Subsequently, the whole carcasses of the 10 selected fishes from each treatment was ground using a grinder (Fleetwood Slicing Machines, Newark, NJ, USA) fitted with a die of 6.35 mm (¼ inch) for analyzing the whole-body composition. All the ground carcasses for each treatment were mixed using a classic series 4.25 L (4.5 quart) tilt-head stand mixer (Kitchen-Aid, Benton Harbor, MI, USA) to form a composite sample. The fish fillet samples were produced by filleting both sides of the fish using a 152.4 mm (6-inch) fillet knife and wearing a cut-resistant glove. These samples were stored at −20 °C until further analysis. The frozen samples were shipped overnight to an AOAC approved commercial laboratory, ATC Scientific (Little Rock, AR, USA) for proximate, fatty acid and amino acid analysis.

2.6. Calculations

The formulas below were utilized for assessing the feed efficiency, growth performance, and body parameters and indexes of the sampled fish across groups [38,39,40].

Weight Gain (WG)

$$\mathrm{WG}=\mathrm{Final} \mathrm{body} \mathrm{weight} \left(\mathrm{g}\right)-\mathrm{Intial} \mathrm{body} \mathrm{weight} \left(\mathrm{g}\right)$$

(2)

Percentage Weight Gain (WG,%)

$$\mathrm{WG}\%={\displaystyle \frac{\mathrm{Final} \mathrm{body} \mathrm{weight} \left(\mathrm{g}\right)-\mathrm{Initial} \mathrm{body} \mathrm{weight} \left(\mathrm{g}\right)}{\mathrm{Initial} \mathrm{body} \mathrm{weight} \left(\mathrm{g}\right)}}\times 100$$

(3)

Specific Growth Rate (SGR, %/day)

$$\mathrm{SGR}={\displaystyle \frac{\ln \mathrm{Final\; body} \mathrm{weight} \left(\mathrm{g}\right) - \ln \mathrm{Initial\; body} \mathrm{weight} \left(\mathrm{g}\right)}{\mathrm{Number} \mathrm{of} \mathrm{days} \mathrm{fed}}}\times 100$$

(4)

Feed Intake (FI, g/fish)

$$\mathrm{FI}={\displaystyle \frac{\mathrm{Total} \mathrm{dry} \mathrm{feed} \mathrm{given} \left(\mathrm{g}\right)}{\mathrm{Number} \mathrm{of} \mathrm{fish}}}$$

(5)

Feed Conversion Ratio (FCR)

$$\mathrm{FCR}={\displaystyle \frac{\mathrm{Feed} \mathrm{consumption} \left(\mathrm{g}\right)}{\mathrm{Body} \mathrm{weight} \mathrm{gain} \left(\mathrm{g}\right)}}$$

(6)

Protein Efficiency Ratio (PER)

$$\mathrm{PER}={\displaystyle \frac{\mathrm{Weight} \mathrm{gain} \left(\mathrm{g}\right)}{\mathrm{Protein} \mathrm{fed} \left(\mathrm{g}\right)}}$$

(7)

Condition Factor (K)

$$\mathrm{K}={\displaystyle \frac{\mathrm{Whole} \mathrm{body} \mathrm{wet} \mathrm{weight} \left(\mathrm{g}\right)}{{\mathrm{Length}}^3}}\times 100$$

(8)

Survival (%)

$$\mathrm{Survival}={\displaystyle \frac{\mathrm{Total} \mathrm{number} \mathrm{of} \mathrm{fish} \mathrm{surviving}}{\mathrm{Total} \mathrm{number} \mathrm{of} \mathrm{fish} \mathrm{stocked}}}\times 100$$

(9)

2.7. Statistical Analysis

The experiment followed a Completely Randomized Design (CRD), where four treatments were assigned randomly to experimental tanks, with each treatment replicated three times, resulting in a total of 12 experimental units. Data were collected from all experimental units following this randomized assignment. Data were analyzed using analysis of variance (ANOVA) to assess treatment effects. When significant differences were detected, Tukey’s Honestly Significant Difference (HSD) test was conducted for multiple comparisons at a significance level of α = 0.05. All statistical analyses were performed using JMP Pro 17 (SAS Institute Inc., Cary, NC, USA). If statistically significant differences were detected (p < 0.05), they were reported as a comparison of letter means, with different superscript letters indicating significant differences and shared superscript letters indicating overlap between groups.

3. Results

3.1. Pellet Floatability and Palatability

There were no significant differences in palatability of any of the 4 feed types among experimental fishes (

Table 3. Feed intake of the 5-day palatability and floatability tests.

Days	Commercial Feed	25% SSBP	50% SSBP	100% SSBP
Day 1	71.67 ± 17.07	69.83 ± 3.17	71.33 ± 1.25	77.67 ± 3.25
Day 2	36.67 ± 3.54	45.50 ± 7.40	56.00 ± 19.75	46.00 ± 9.53
Day 3	47.17 ± 5.48	56.16 ± 9.69	49.00 ± 2.78	50.83 ± 3.68
Day 4	48.50 ± 1.32	54.50 ± 9.83	49.50 ± 2.29	47.83 ± 11.89
Day 5	61.83 ± 7.52	61.00 ± 5.39	49.67 ± 7.18	53.50 ± 5.30
% Pellet Float	100 ± 0.00	100 ± 0.00	97.33 ± 2.19	99.00 ± 0.58

The values are Means ± Std. Dev; no statistically significant differences (p > 0.05) were detected among group means.

). Fish in all 12 tanks ate until apparent satiation twice a day for the 5 days of the palatability study. Fish ate more on the first day at a range of approximately 70–77 g compared to all other days, likely due to that being the first day being fed after a 2 day fast. Their feed intake decreased in subsequent days which ranged approximately from 35–60 g, but no differences were exhibited between groups. In addition, there were no significant differences in pellet floatability with a >95% float rate being recorded for all pellets and a 100% float rate being recorded for half of the sampled feeds.

3.2. Feeding Trials

Feeding trial statistics and data are shown in

Table 4. Changes in weight and lengths of juvenile domesticated striped bass (Morone saxatilis) for weight and length every 30 days during the course of study.

	Commercial Feed	25% SSBP	50% SSBP	100% SSBP
W1 (g)	38.23 ± 8.86 A	35.50 ± 9.26 A	32.08 ± 11.51 A	35.88 ± 13.58 A
W2 (g)	60.77 ± 17.54 A	56.45 ± 10.90 A	55.40 ± 16.82 A	62.75 ± 14.29 A
W3 (g)	86.10 ± 25.12 A	79.45 ± 21.23 A	85.20 ± 25.41 A	80.13 ± 24.33 A
W4 (g)	129.35 ± 35.86 A	107.33 ± 30.31 AB	114.41 ± 41.13 AB	103.90 ± 24.02 B
W5 (g)	183.12 ± 54.71 A	154.62 ± 51.25 AB	157.55 ± 42.74 AB	136.23 ± 34.50 B
W6 (g)	217.52 ± 35.30 A	218.78 ± 11.08 A	212.23 ± 36.37 A	197.70 ± 27.59 A
L1 (mm)	145.03 ± 11.62 A	141.17 ± 12.55 AB	134.37 ± 14.02 B	140.20 ± 14.51 AB
L2 (mm)	161.47 ± 13.72 A	157.33 ± 10.20 A	154.97 ± 14.46 A	160.17 ± 13.16 A
L3 (mm)	182.77 ± 16.36 A	175.30 ± 14.68 A	179.53 ± 17.02 A	174.57 ± 18.08 A
L4 (mm)	206.27 ± 17.20 A	194.90 ± 17.44 AB	196.70 ± 21.81 AB	192.77 ± 16.38 B
L5 (mm)	228.13 ± 23.15 A	214.90 ± 23.80 AB	217.37 ± 18.19 AB	207.57 ± 14.48 B
L6 (mm)	239.77 ± 7.13 A	241.23 ± 4.70 A	234.77 ± 12.19 A	232.97 ± 8.49 A

Means ± Std Dev ^A with a letter in common within the same row are not significantly different (p > 0.05). W = Weight, L = Length, W₁/L₁ = First sampling (base line); W₂/L₂ = Second sampling; W₃/L₃ = Third sampling; W₄/L₄ = Fourth sampling; W₅/L₅ = Fifth sampling; W₆/L₆ = Sixth sampling.

and

Table 5. Effect of feed type on WG, SGR, FI, FCR, PER, K, and survival in juvenile domesticated striped bass (Morone saxatilis).

	Commercial Feed	25% SSBP	50% SSBP	100% SSBP
WG (g)	179.28 ± 57.23	183.28 ± 50.74	180.15 ± 74.15	161.82 ± 60.45
WG%	486.68 ± 177.46	556.05 ± 216.44	655.49 ± 437.06	499.36 ± 227.58
SGR	1.15 ± 0.22	1.22 ± 0.23	1.26 ± 0.34	1.14 ± 0.28
FI (g)	240.58 ± 21.98	239.86 ± 6.55	230.04 ± 15.74	230.66 ± 25.20
FCR	1.36 ± 0.13	1.31 ± 0.12	1.31 ± 0.21	1.44 ± 0.14
PER	1.65 ± 0.16	1.70 ± 0.16	1.73 ± 0.30	1.55 ± 0.15
K	1.24 ± 0.02	1.24 ± 0.11	1.29 ± 0.18	1.27 ± 0.16
% Survival	87.33 ± 2.30	93.33 ± 1.15	89.33 ± 4.16	89.33 ± 10.26

The values are Means ± Std Dev, no statistically significant differences (p > 0.05) were detected among group means. SSBP = Smoked salmon by-product, WG = weight gain, WG% = weight gain percentage, SGR = specific growth rate, FI = feed intake, FCR = feed conversion ratio, PER = protein efficiency ratio, and K = condition factor.

. Significant differences were found between weights and lengths of sampled fish during two sampling periods (week 12 and week 16) between the commercial feed and the 100% SSBP feed. During the fourth sampling, juvenile domesticated striped bass fed the commercial diet had a respective weight and length of 129.35 ± 35.86 g and 206.27 ± 17.20 mm, whereas striped bass fed the 100% SSBP had a weight and length of 103.90 ± 24.0 g and 192.77 ± 16.38 mm. During the fifth sampling, striped bass fed the commercial diet had a respective weight and length of 183.12 ± 54.71 g and 206.27 ± 17.20 mm, while striped bass fed 100% SSBP feed had a weight and length of 228.13 ± 23.15 g and 207.57 ± 14.48 mm. For the final sampling period (week 20), statistically significant differences in weight and lengths were no longer detected. Wide standard deviation can be seen across sampling groups in both weight and length, with the more drastic standard deviations being recorded in weight. These large standard deviations were an observed trend across groups and may correlate to the Mycobacterium infection of fish within the system.

There were no significant differences in weight gain (p = 0.5232), percent weight gain (p = 0.0908), specific growth rate (p = 0.2568), feed intake (p = 0.8398), FCR (p = 0.7098), PER (p = 0.7153), or condition factor (p = 0.5554) across groups. For FCR, the 25% SSBP and 50% SSBP based feeds were the most efficient at an average of 1.31 ± 0.12 and 1.31 ± 0.21, respectively; however, for the 100% SSBP feed, PER was the lowest at 1.55 ± 0.15. The 50% SSBP feed had the highest PER at 1.73 ± 0.30, although the differences were statistically non-significant across the treatments for FCR and PER. All tanks averaged approximately 87–93% survivability with no statistically significant differences. However, in one tank (100% SSBP) there was an outlier of 18 mortalities whereas most tanks usually suffered 5–6. It is not possible to say what this spike was attributable to. Additionally, two mortalities were found within the sump of the RAS system, as such it was not possible during sampling to determine which tanks the fish came from.

3.3. Composition of the Fish After the Feeding Trials

Table 6. Proximate composition (dry basis) of juvenile domesticated striped bass (fillets and whole fish) at the end of feeding trials.

	Diet Type	Crude Protein (%)	Crude Fat by Acid Hydrolysis (%)	Crude Fiber (%)	Ash (%)	Sodium (%)
Fillet	0% SSBP	70.59 ± 2.98 A	25.63 ± 3.42 D	2.94 ± 0.64 A	5.60 ± 0.29 B	0.25 ± 0.01 C
	25% SSBP	62.46 ± 4.13 C	34.66 ± 2.82 BCD	2.00 ± 0.72 A	5.79 ± 2.76 B	0.25 ± 0.01 C
	50% SSBP	62.86 ± 1.57 BC	29.91 ± 2.47 CD	2.24 ± 0.54 A	5.86 ± 1.84 B	0.28 ± 0.04 BC
	100% SSBP	69.94 ± 2.75 AB	27.27 ± 1.53 CD	2.77 ± 1.02 A	6.16 ± 1.51 B	0.28 ± 0.01 BC
Whole Fish	0% SSBP	44.47 ± 1.82 D	42.17 ± 8.56 ABC	1.92 ± 0.21 A	9.23 ± 1.29 AB	0.32 ± 0.03 ABC
	25% SSBP	42.44 ± 3.48 D	37.50 ± 1.44 BCD	1.52 ± 0.89 A	9.82 ± 2.18 AB	0.29 ± 0.02 BC
	50% SSBP	49.37 ± 4.93 D	48.51 ± 1.24 AB	1.92 ± 0.13 A	11.01 ± 0.16 AB	0.34 ± 0.01 AB
	100% SSBP	48.66 ± 1.73 D	56.56 ± 4.60 A	2.47 ± 0.86 A	12.64 ± 0.69 A	0.39 ± 0.02 A

Means ± Std Dev ^A represents significant differences between values in a column and not sharing the same letter.

shows the proximate composition of the juvenile domesticated striped bass (fillets and whole fish) at the end of the feeding trials. The mean crude protein content in fillets ranged from 70.59 ± 2.98% to 62.46 ± 4.13% with higher protein content being found in fillets of fish that were fed 0% SSBP feeds and the lowest was found in fish fed with 25% SSBP. Significant differences (p < 0.05) were observed between 0% SSBP and fishes fed either 25% or 50% SSBP. Whole fish having the highest crude protein content of 49.37 ± 4.93% were found in fish that were fed 50% SSBP and the lowest of 42.44 ± 3.48% were in fish fed with 25% SSBP feeds. However, no significant differences were observed amongst the treatments. The mean crude fat in fish fillets ranged from 25.63 ± 3.42% to 34.66 ± 2.82% in fishes fed 0% SSBP and 25% SSBP, respectively. Significant differences (p < 0.05) were found between these treatments, while for the whole fish, the mean crude fat content ranged from 37.50 ± 1.44% to 56.56 ± 4.60% for 25% SSBP- and 100% SSBP-fed fishes, respectively. Significant differences were found between these treatments. No significant differences were found in either the fillets or whole fish for crude fiber, ash, and sodium (fillets only). There was significant difference between the 100% SSBP and 25% SSBP fed whole fish.

The fatty acid profile of the fish fillets and whole fish is presented in

Table 7. Fatty acid profile of fillet and whole fish of juvenile domesticated striped bass at the end of feeding trials.

Fatty Acids	0% SSBP		25% SSBP		50% SSBP		100% SSBP
	Fillet	Whole Fish	Fillet	Whole Fish	Fillet	Whole Fish	Fillet	Whole Fish
Myristic (C 14:0)	4.04 ± 0.01 A	4.08 ± 0.01 A	3.60 ± 0.01 C	3.86 ± 0.04 B	3.31 ± 0.00 D	3.80 ± 0.04 B	2.62 ± 0.02 F	2.91 ± 0.01 E
Palmitic (C 16:0)	22.14 ± 0.05 C	21.78 ± 0.04 CD	21.94 ± 0.05 CD	22.82 ± 0.01 B	21.69 ± 0.00 D	24.77 ± 0.35 A	20.63 ± 0.01 E	23.04 ± 0.01 B
Palmitoleic (C 16:1)	9.95 ± 0.01 B	10.21 ± 0.05 A	9.00 ± 0.07 D	9.15 ± 0.05 D	8.54 ± 0.00 E	9.45 ± 0.12 C	6.54 ± 0.03 G	7.00 ± 0.01 F
Stearic (C 18:0)	3.22 ± 0.01 A	3.09 ± 0.01 B	3.19 ± 0.01 A	3.11 ± 0.01 B	2.91 ± 0.00 C	2.83 ± 0.04 DE	2.81 ± 0.00 E	2.89 ± 0.01 CD
Oleic (C 18:1)	30.73 ± 0.13 H	31.56 ± 0.03 G	33.72 ± 0.06 E	33.02 ± 0.18 F	34.89 ± 0.00 D	36.42 ± 0.11 C	38.98 ± 0.10 B	39.75 ± 0.02 A
Linoleic (C 18:2)	10.90 ± 0.03 H	11.02 ± 0.02 G	11.63 ± 0.01 F	11.94 ± 0.01 E	13.13 ± 0.00 D	14.58 ± 0.01 C	15.70 ± 0.06 B	17.69 ± 0.00 A
alpha-Linolenic (C 18:3 n-3)	1.42 ± 0.03 F	1.42 ± 0.00 F	1.55 ± 0.00 E	1.54 ± 0.00 E	1.69 ± 0.00 D	1.83 ± 0.00 C	2.01 ± 0.00 B	2.13 ± 0.01 A
Gondoic (C 20:1)	1.67 ± 0.11 D	1.78 ± 0.01 CD	1.72 ± 0.08 CD	1.95 ± 0.01 BC	2.04 ± 0.00 B	1.05 ± 0.10 E	2.48 ± 0.02 A	1.24 ± 0.02 E
Arachidonic (C 20:4 n-6)	0.69 ± 0.01 A	0.10 ± 0.00 D	0.10 ± 0.01 D	0.12 ± 0.01 D	0.11 ± 0.00 D	0.33 ± 0.03 B	0.15 ± 0.01 D	0.21 ± 0.01 C
Eicosapentaenoic (EPA) (C 20:5 n-3)	5.12 ± 0.00 A	4.87 ± 0.01 B	4.28 ± 0.00 C	3.69 ± 0.02 D	3.29 ± 0.00 E	1.66 ± 0.09 G	1.82 ± 0.01 F	0.67 ± 0.01 H
Docosahexaenoic (DHA) (C 22:6 n-3)	5.31 ± 0.01 A	4.65 ± 0.01 B	4.60 ± 0.01 B	3.77 ± 0.02 C	3.45 ± 0.00 D	0.82 ± 0.11 F	2.12 ± 0.02 E	0.41 ± 0.01 G
Trans Fats	<0.01	<0.01	<0.01	<0.01	<0.01	<0.01	<0.01	<0.01
Saturated Fats	35.89 ± 0.73 A	32.37 ± 0.90 AB	31.36 ± 0.90 B	32.87 ± 0.90 AB	30.78 ± 0.90 B	32.37 ± 0.90 AB	28.18 ± 0.90 B	29.44 ± 0.90 B
Monounsaturated	42.78 ± 0.22 F	43.98 ± 0.04 E	44.84 ± 0.06 D	44.51 ± 0.14 D	45.84 ± 0.00 C	47.30 ± 0.07 B	48.29 ± 0.12 A	48.26 ± 0.00 A
Polyunsaturated	21.33 ± 2.92 A	23.66 ± 0.04 A	23.81 ± 0.01 A	22.62 ± 0.07 A	23.38 ± 0.00 A	20.33 ± 0.29 A	23.54 ± 0.09 A	22.30 ± 0.00 A
Total omega-3	8.54 ± 2.93 AB	10.94 ± 0.01 A	10.44 ± 0.01 A	9.00 ± 0.00 AB	8.44 ± 0.00 ABC	4.38 ± 0.21 BC	5.96 ± 0.01 ABC	3.28 ± 0.01 C
Total omega-6	12.45 ± 0.03 G	11.88 ± 0.01 H	12.65 ± 0.01 F	12.95 ± 0.04 E	14.31 ± 0.00 D	15.69 ± 0.09 C	17.14 ± 0.05 B	18.80 ± 0.01 A
Total omega-9	32.43 ± 0.21 E	33.37 ± 0.02 D	35.47 ± 0.13 C	35.00 ± 0.19 C	36.96 ± 0.00 B	37.50 ± 0.21 B	41.49 ± 0.08 A	41.01 ± 0.00 A

Means ± Std Dev ^A represents significant differences (p < 0.05) between values in a row and not sharing the same letter. All the values in this table are expressed in percentages.

. It can be observed that in the omega-3 fatty acids of interest, the DHA (2.12 ± 0.02% to 5.31 ± 0.01%) and EPA (1.82 ± 0.01% to 5.12 ± 0.00%) components are higher than alpha-Linolenic acid (1.42 ± 0.04% to 2.01 ± 0.00%) in fillets. A similar trend was observed with whole-fish fractions as well but their DHA (0.41 ± 0.01% to 4.65 ± 0.01%) and EPA (0.67 ± 0.01% to 4.87 ± 0.01%) levels were lower than those of the fillets, while the mean alpha-Linolenic acid was higher (1.42 ± 0.04% to 2.01 ± 0.00%). The DHA and EPA in both cases decreased as the addition of SSBP increased in the feeds. The reverse trend was observed for alpha-Linolenic acid. The monounsaturated fats contributed the highest fraction followed by saturated fats and polyunsatuared fats in both fillets and whole fish. Within the unsaturated fats, polyunsaturated fats were highest with omega-9 being the biggest component followed by omega-6, whereas the monounsaturated fat (omega-3) was found to be the least.

The protein quality of the fillet and whole fish can be determined by the amino acid profile (

Table 8. Amino acid profile of fillet and whole fish of juvenile domesticated striped bass at the end of feeding trials.

Amino Acids	0% SSBP		25% SSBP		50% SSBP		100% SSBP
	Fillet	Whole Fish	Fillet	Whole Fish	Fillet	Whole Fish	Fillet	Whole Fish
Alanine	1.27 ± 0.01 A	1.14 ± 0.03 AB	1.13 ± 0.06 AB	0.96 ± 0.03 B	1.18 ± 0.18 AB	0.99 ± 0.09 AB	1.26 ± 0.06 A	0.99 ± 0.16 AB
Arginine	0.86 ± 0.00 AB	0.78 ± 0.12 B	0.76 ± 0.01 B	0.72 ± 0.12 B	0.85 ± 0.09 AB	0.83 ± 0.08 AB	0.99 ± 0.01 A	0.75 ± 0.07 B
Aspartic Acid	1.83 ± 0.04 AB	1.35 ± 0.13 CD	1.76 ± 0.12 AB	1.20 ± 0.13 D	1.54 ± 0.13 BC	1.37 ± 0.09 CD	1.87 ± 0.08 A	1.25 ± 0.17 CD
Cysteine	0.19 ± 0.00 A	0.12 ± 0.00 B	0.19 ± 0.00 A	0.10 ± 0.00 B	0.17 ± 0.00 A	0.13 ± 0.00 B	0.20 ± 0.00 A	0.12 ± 0.00 B
Glutamic Acid	2.99 ± 0.09 A	2.02 ± 0.11 B	2.70 ± 0.05 A	1.67 ± 0.11 B	2.57 ± 0.30 A	2.05 ± 0.10 B	2.70 ± 0.03 A	1.78 ± 0.32 B
Glycine	0.91 ± 0.01 A	1.10 ± 0.03 A	0.99 ± 0.06 A	0.85 ± 0.03 A	1.01 ± 0.15 A	0.88 ± 0.05 A	0.93 ± 0.04 A	1.01 ± 0.03 A
Histidine	0.28 ± 0.01 B	0.30 ± 0.06 B	0.27 ± 0.01 B	0.28 ± 0.06 B	0.35 ± 0.02 AB	0.31 ± 0.01 B	0.46 ± 0.07 A	0.33 ± 0.11 AB
Isoleucine	0.95 ± 0.05 A	0.67 ± 0.05 B	0.93 ± 0.02 A	0.66 ± 0.05 B	0.85 ± 0.12 A	0.61 ± 0.06 B	0.91 ± 0.06 A	0.62 ± 0.05 B
Leucine	1.48 ± 0.12 A	1.04 ± 0.05 B	1.51 ± 0.04 A	1.08 ± 0.05 B	1.46 ± 0.18 A	0.94 ± 0.06 B	1.47 ± 0.04 A	1.02 ± 0.04 B
Lysine	2.00 ± 0.06 A	1.25 ± 0.09 B	1.94 ± 0.02 A	1.32 ± 0.09 B	1.79 ± 0.23 A	1.11 ± 0.07 B	1.87 ± 0.05 A	1.25 ± 0.06 B
Methionine	0.55 ± 0.01 A	0.44 ± 0.01 B	0.57 ± 0.01 A	0.33 ± 0.01 C	0.55 ± 0.01 A	0.47 ± 0.01 B	0.57 ± 0.01 A	0.37 ± 0.01 C
Ornithine	0.16 ± 0.03 AB	0.14 ± 0.04 AB	0.17 ± 0.02 A	0.12 ± 0.04 AB	0.19 ± 0.07 A	0.07 ± 0.01 B	0.11 ± 0.03 AB	0.09 ± 0.02 AB
Phenylalanine	0.72 ± 0.04 A	0.58 ± 0.04 B	0.71 ± 0.02 A	0.54 ± 0.04 B	0.76 ± 0.10 A	0.48 ± 0.04 B	0.72 ± 0.01 A	0.47 ± 0.02 B
Proline	0.59 ± 0.00 A	0.73 ± 0.05 A	0.61 ± 0.00 A	0.53 ± 0.05 A	0.60 ± 0.09 A	0.61 ± 0.06 A	0.72 ± 0.06 A	0.63 ± 0.05 A
Serine	0.57 ± 0.01 B	0.59 ± 0.04 B	0.55 ± 0.05 BC	0.44 ± 0.04 C	0.58 ± 0.04 B	0.57 ± 0.04 B	0.74 ± 0.03 A	0.47 ± 0.07 BC
Taurine	0.29 ± 0.02 A	0.28 ± 0.02 A	0.28 ± 0.03 A	0.26 ± 0.02 A	0.22 ± 0.03 A	0.27 ± 0.03 A	0.30 ± 0.04 A	0.25 ± 0.04 A
Threonine	0.72 ± 0.05 ABCD	0.68 ± 0.04 BCD	0.84 ± 0.05 A	0.63 ± 0.04 CD	0.73 ± 0.09 ABC	0.62 ± 0.03 CD	0.82 ± 0.03 AB	0.58 ± 0.08 D
Tryptophan	0.20 ± 0.00 AB	0.14 ± 0.00 C	0.15 ± 0.00 BC	0.09 ± 0.00 D	0.11 ± 0.00 CD	0.08 ± 0.00 D	0.21 ± 0.00 A	0.13 ± 0.00 CD

Means ± Std Dev ^A represents significant differences (p < 0.05) between values in a row and not sharing the same letter. All the values in this table are in percentages.

). The table shows that the fish fillets and whole fish were good sources of essential amino acids. Overall, the total mean amino acid content was found to be higher in fillet as compared to whole fish across all treatments. Glutamic acid was found with the highest concentration, ranging from 2.57 ± 0.3% to 2.99 ± 0.09% in 50% SSBP fed fillets and 0% SSBP fillets, respectively. The levels of glutamic acid were lower in whole fish and ranged from 1.67 ± 0.11% to 2.05 ± 0.10% in 25% SSBP fed fishes and 50% SSBP fed fishes, respectively. Amongst the essential amino acids, lysine concentration was found to be the highest in both fillets and whole fish. The lysine levels in fillets were significantly higher (p < 0.0001) that that in whole fish for all feed types. The highest level of lysine was in fillets and ranged from 1.79 ± 0.23% to 2.00 ± 0.06% in 50% SSBP and 0% SSBP, respectively. The concentration of lysine in whole fish was found to be the highest in 25% SSBP treatment (1.32 ± 0.09%) and lowest in 50% SSBP treatment (1.11 ± 0.07%). The methionine levels also followed the same trend, with significantly higher levels (p < 0.0001) in fillets for all treatment groups as compared to whole fish. Therefore, the average concentration of essential amino acids in whole fish (across all treatments) is somewhat lower than that of fish fillet.

3.4. Accelerated Shelf-Life Study

For the baseline start (week 0) of the 12-week accelerated shield life study, SSBP FM was reported to have the final values of 3.70 ± 0.06% FFA, 14.00 ± 0.00 meq/kg oil PV, 8.33 ± 0.23 p-AV, and a TOTOX of 15.73 ± 0.14 (

Table 9. Accelerated shelf-life evaluation of SSBP FM and FO.

	Parameters	Week 0	Week 2	Week 4	Week 6	Week 8	Week 10	Week 12
FM	FFA (%)	3.70 ± 0.06 B	0.55 ± 0.04 G	1.61 ± 0.05 F	2.22 ± 0.05 E	2.67 ± 0.11 D	4.62 ± 0.08 A	3.02 ± 0.13 C
	PV (meq/kg)	14.00 ± 0.00 A	13.00 ± 0.00 B	10.67 ± 0.58 C	6.50 ± 0.35 E	8.70 ± 0.40 D	7.27 ± 0.31 E	11.00 ± 0.00 C
	p-AV	8.33 ± 0.23 D	14.2 ± 0.70 C	9.83 ± 0.95 CD	74.53 ± 2.32 B	86.87 ± 2.67 A	9.47 ± 0.21 D	73.40 ± 2.23 B
	TOTOX	36.33 ± 0.23 D	40.2 ± 0.70 D	31.17 ± 0.31 E	87.53 ± 2.25 C	104.27 ± 3.46 A	24.00 ± 0.82 F	95.40 ± 2.23 B
FO	FFA (%)	1.60 ± 0.00 AB	1.50 ± 0.00 BC	1.70 ± 0.00 A	1.47 ± 0.00 C	1.47 ± 0.00 C	1.60 ± 0.00 AB	1.67 ± 0.06 A
	PV (meq/kg)	21.00 ± 0.00 A	18.00 ± 0.00 C	14.33 ± 0.58 D	19.00 ± 0.00 B	14.00 ± 0.00 D	19.33 ± 0.58 B	8.70 ± 0.10 E
	p-AV	7.83 ± 0.47 E	12.30 ± 0.70 D	13.9 ± 0.20 C	24.23 ± 0.21 B	24.60 ± 0.30 B	27.10 ± 0.17 A	27.43 ± 0.49 A
	TOTOX	49.83 ± 0.47 D	48.3 ± 0.70 D	42.57 ± 0.99 F	62.23 ± 0.21 B	52.60 ± 0.30 C	65.77 ± 1.33 A	44.83 ± 0.68 E

Means ± Std Dev ^A represents significant differences (p < 0.05) between values in a row and not sharing the same letter. Samples were collected and analyzed every two weeks for twelve weeks. FM = smoked salmon by-product fish meal; FO = smoked salmon by-product fish oil; p-AV = p-Ansidine value; FFA = free fatty acid; PV = peroxide value; TOTOX = total oxidation value.

). SSBP FO was reported to have the final values of 1.60 ± 0.00% FFA, 21.00 ± 0.00 meq/kg oil PV, 7.83 ± 0.47 p-AV, and a TOTOX of 49.83 ± 0.47. Six weeks into the study, a slight decrease can be observed in both FFA and PV in both SSBP FM and FO; however, there was a substantial increase in p-AV and subsequently TOTOX at this time with them being 74.53 ± 2.32 p-AV and 87.53 ± 2.25 TOTX for SSBP FM. For SSBP FO during this week, these values were 24.23 ± 0.21 p-AV and 62.23 ± 0.21 TOTOX. At the end of the 12-week accelerated shelf-life study, SSBP FM was reported to have final values of 3.02 ± 0.13% FFA, 11.00 ± 0.00 meq/kg oil PV, 73.40 ± 2.23 p-AV, and a TOTOX of 95.40 ± 2.23. SSBP FO was reported to have the final values of 1.67 ± 0.06% FFA, 8.70 ± 0.10 meq/kg oil PV, 27.43 ± 0.49 p-AV, and a TOTOX of 44.83 ± 0.68. The data seemingly never showed a consistent trend with values dropping and then rising in subsequent weeks throughout the trial.

4. Discussion

The buoyancy of fish feeds is mainly determined by the fish species that is being fed. Extrusion is used for making floating feeds and the extrusion process or pellet milling can also be used to make sinking feeds. The extruded feeds used in this study had good floatability with a value greater than 97%. This is a good indicator of proper feed formulation as well as skillful process control. All the feeds used in the 5-day palatability study were accepted by the fish. This meant that the FM and FO produced from SSBP were similar in palatability to the commercial FM and FO used in the study. The gustatory system of fish is sensitive to the palatability of feeds and directly impacts the feed intake, satiation, resultant feed waste, and growth. The chemosensory system of the fish easily recognizes and locates the feed, supporting the action of consumption and ingestion, especially when the ingredients are derived from marine animal origins like FM, krill meal, etc. [41]. However, it should be noted that sensory perception to locate and consume feed by the fish is short-term (hours and days) activity, whereas in the long term (weeks and months), the feed intake in aquatic animals is governed by the energy or protein demand and the animals’ endocrine regulatory processes that dominate the control over appetite [3].

Weight and length performance of juvenile domesticated striped bass over the 20-week study period did not show any negative dietary treatment effects on the fish weights. At up to 8 weeks, there was no significant difference in the weights of the fish across all treatments. However, after that, significant differences in weights (p = 0.0198 in 12th week and p = 0.0022 in 16th week) became apparent amongst the treatments. This might have been due to the dietary effect of lower amino acid level/concentration in the SSBP FM as compared to commercial FM. The fish fed with 100% SSBP FM showed the least weight gain after 16 weeks; however, at 20 weeks, statistically significant differences were no longer detected among diets for this variable. A study by Baek and Cho [42] showed that 40% FM could be replaced by tuna by-product meal for red seabream (Pagrus major) without compromising growth, feed consumption, and feed utilization. In another study, it was found that 100% replacement of FM with shrimp by-product meal in feeds for the omnivorous Nile tilapia (Oreochromis niloticus) had highest final length, length gain and other growth performance parameters [43]. The lengths of the fish in our study showed less variation than the weights although statistically significant differences were still found between the lengths of fish fed the commercial diet and 100% SSBP fed fish in weeks 12 (p = 0.0270) and 16 (p = 0.0019). These differences were not statistically retained in the final sampling that occurred at week 20. According to Voorhees et al. [44], the length of the fish can be influenced by factors such as feeding rates, rearing densities, dietary nutritional differences, environmental stress, and fish health. Notwithstanding the impact of mycobacteriosis, there have been some variations in the growth performance of juvenile domesticated striped bass due to its short history of domestication. This study used progeny from the seventh-generation striped bass spawned at the PAFL research center. Andersen et al. [28] reported that the fourth-generation had higher weight gain than what was found for the second generation. This along with the standard deviations of the weights during sampling, are evident of variations expected in growth performance between generations. Patino [29] reported the variations in weight and length of juvenile domesticated striped bass when fed with soybean meals of varying fatty acid composition. Although the fish diets were different in the inclusion levels of SSBP FM, significant differences in 25% SSBP and 50% SSBP diets when compared to commercial feed were not found in the length or weight of the fish during any measured time-period. This suggests that lower levels of substitution with SSBP FM and FO do not negatively affect the length of the fish.

When compared to a commercial Pacific whiting 68% FM, the SSBP FM had a lower quality amino acid profile (

Table 10. Amino acid profile comparison of commercially available Pacific WFM and SSBP FM.

Amino Acid	Pacific WFM (%)	SSBP FM (%)
Alanine	4.0	1.62 ± 0.03
Arginine	4.5	1.67 ± 0.03
Aspartic acid	5.3	2.36 ± 0.10
Cystine	0.5	0.23 ± 0.02
Glutamic acid	8.0	3.19 ± 0.32
Glycine	5.4	2.25 ± 0.28
Histidine	1.4	0.68 ± 0.02
Isoleucine	2.6	1.06 ± 0.12
Leucine	4.9	1.43 ± 0.50
Lysine	5.0	1.78 ± 0.13
Methionine	1.6	0.78 ± 0.02
Phenylalanine	2.7	0.97 ± 0.06
Proline	3.6	1.29 ± 0.07
Serine	3.0	1.12 ± 0.06
Threonine	2.0	1.11 ± 0.04
Tryptophan	0.5	0.37 ± 0.02
Tyrosine	2.1	0.93 ± 0.09
Valine	3.2	1.14 ± 0.06

WFM data were acquired from BioOregon Protein, Inc., Warrenton, OR, USA. SSBP FM values are presented as Means ± Std. Dev. WFM = whitefish meal, and SSBP FM = smoked salmon by-product fish meal.

). The commercial FM was higher in 10 essential amino acids which have been reported by Xing et al. [45] as arginine, histidine, isoleucine, leucine, valine, lysine, sulfur amino acids (methionine + cysteine), total aromatic amino acids (phenylalanine + tyrosine), threonine, and tryptophan. While the crude protein content of both FMs were 74% (dry basis) [34], these lower values might have been due to the addition of a higher proportion of skins than the trimmings (2:1 ratio of skins/trimmings) when producing the FM. Therefore, the substitution of SSBP FM at higher levels may enable a poorer nutritional profile of feeds as compared to commercial FM and this may affect the growth of fish. This was also observed in this study.

On comparing the SSBP FO to a commercial Pacific whiting FO (

Table 11. Fatty acid profile comparison of commercially available Pacific WFO, SSBP FO, and reports from farmed salmon.

Fatty Acids	Pacific WFO (%)	SSBP FO (%)	Farmed Salmon Reports (%)
Myristic (C 14:0)	3.3	2.01 ± 0.12	1.5–5.5
Palmitic (C 16:0)	21.8	12.06 ± 0.01	6.5–12.0
Stearic (C 18:0)	3.5	3.97 ± 0.09	2.0–5.0
Total SFA	30.0	21.29 ± 0.36	X
Palmitoleic (C 16:1 n-7)	7.3	2.33 ± 0.04	2.0–5.0
Erucic (C 22:1 n-9)	1.0	3.86 ± 0.10	3.0–7.0
cis-Oleic (C 18:1 n-9)	28.7	36.41 ± 0.76	30.0–47.0
Total MUFA	38.5	46.86 ± 0.59	X
alpha-Linolenic (C 18:3 n-3)	0.8	4.19 ± 0.06	3.0–6.0
cis-11,14,17-Eicosatrienoic (C 20:3 n-3)	0.2	0.34 ± 0.05	X
Docosahexaenoic (DHA) (C 22:6 n-3)	9.9	4.67 ± 0.29	3.0–10.0
Eicosapentaenoic (EPA) (C 20:5 n-3)	14.6	0.36 ± 0.01	2.0–6.0
Total n-3 PUFA	26.6	9.46 ± 0.42	X
Arachidonic (C 20:4 n-6)	1.1	0.35 ± 0.04	ND–1.2
Linoleic (C 18:2 n-6)	X	20.06 ± 0.13	8.0–15.0
Total n-6 PUFA	2.8	22.12 ± 0.11	X
Total PUFA	29.5	31.86 ± 0.27	X

WFO data were acquired from BioOregon Protein. Farmed salmon reports are from the FAO and are noted as a range of values. SSBP FO values are presented as means ± std dev. WFO = whitefish oil; SSBP FO = smoked salmon by-product fish oil; X = not analyzed; ND = not detected; SFA = saturated fatty acids; MUFA = monounsaturated fatty acids; PUFA = polyunsaturated fatty acids; n-3 = omega-3 fatty acids; n-6 = omega-6 fatty acids

), it was found that the SSBP FO had higher monounsaturated fats and omega-6 polyunsaturated fats, whereas the commercial sample had higher levels of saturated fatty acids, and omega-3 polyunsaturated fatty acids. The fatty acid composition depends on the type of feed that the fish consumes. The salmon by-products were obtained from farmed salmon, which may have feeds with higher plant-based feeds and resulting in derived oils containing higher ratios of omega-6/omega-3 content. Jensen et al. [46] reported that farmed salmons had an omega-6 to omega-3 ratio of more than 10 times that of wild salmon (0.7 vs. 0.05). They further reported that the high content of linoleic and alpha-linolenic acid in farmed salmon shows the substantial inclusion of vegetable oils in the feed. This has occurred due to changes in the price and availability of marine resources [47].

There was no significant difference in the weight gain of juvenile domesticated striped bass after 20 weeks of feeding trials. This was possibly due to the high variation in weights (observed as large standard deviations) across all treatments. A study conducted by Patino [29] in the same facility using the same fish species had much smaller variations in weight within each group of treatments. The larger variations may have been due to the impact of mycobacteriosis stressing the physiology of fish from the start of the study. Additionally, there was an inverse relationship regarding growth as fish weights declined relative to increased inclusion levels of FM and FO from the by-products. Proteins constitute the single largest and the most expensive ingredient category in feed formulations, responsible for the growth and health of the fish [48]. Protein provides essential and non-essential amino acids, which are critical for muscle formation and other functions in fish [49]. There was a difference in the various amino acid levels between the commercial Pacific whiting FM and the SSBP FM (Table 10). FM from trimmings have a different nutritional value that is high in mineral content but lower in protein value as compared to conventional FM [50,51]. Sulfur-containing amino acids are the most limiting in practical diet formulations for hybrid striped bass [32]. Several studies have shown that dietary methionine deficiency has led to poor growth and feed efficiency in juvenile Jian carp (Cyprinus carpio var. Jian) [52], fingerling rohu (Labeo rohita) [53], juvenile Cobia (Rachycentron canadum) [54], juvenile hybrid striped bass (Morone chrysops x M. saxatilis) [55], and juvenile European sea bass (Dicentrarchus labrax) [5]. The sulfur-containing amino acid levels were much higher in the commercial FM as compared to the by-product FM, which might have depressed the growth outcome seen in juvenile domesticated striped bass fed the test diets.

The weight gain percent also did not have a significant difference amongst the treatments. A major reason was the high degree of standard deviation associated within each group. However, the lowest weight gain was observed in the fish that were fed diets containing 100% SSBP FM and FO. The trend was similar to that observed in percent weight gain as well. The 50% by-product FM and FO inclusion group showed the highest percent weight gain, but it also had the highest standard deviation. This indicates the large differences observed in the weights of fish that were sampled randomly throughout the study. The SGR did not show any significant difference amongst the treatments. SGR is a growth metric that is calculated to understand the growth of fish over a small period of time where the weight of the fish, especially small fish, increases exponentially [56]. The lowest SGR was observed in fish fed with 100% SSBP FM and FO, whereas the highest was found in the fish fed with diets containing commercial FM and FO. As discussed earlier, the difference in the amino acid profile of both meals may have impacted this outcome. Poorer amino acid profile of FM from the by-product led to lower weight gain and thus lower SGR. However, a simple solution to this can be obtained by providing an amino acid supplement in the diet.

The FI was not significantly different amongst the different treatments. The FCR was shown to be greater in fishes fed 100% SSBP FM and FO based feeds diets when compared to 100% commercial diets or supplemented SSBP diets. The span of difference in weight gain between these two treatments as compared to the treatments with 25% and 50% SSBP inclusion might have been the cause. The results also show that FCR was affected by weight gain, which is dependent on the diet nutritional quality, especially the amino acid profile of the FM. The PER also showed an inverse trend to FCR with greater efficiency in the control feed and supplemental inclusion feeds as opposed to feed with 100% substitution of commercial FM and FO. Protein efficiency ratio (PER) is the easiest method of assessing the quality of proteins [57]. The major reason for such a result is due to the difference in weights of these fish in these groups, as well as the mortality of these groups. As mortality was observed, the fewer fish would attain more food due to reduced in-system competition. As the feeds were isonitrogenous, the feeds had approximately the same 45% targeted protein content, but lower weights of the fish in the 100% substitution group lowered the PER values. This would indicate that the feed protein was not as efficient in increasing the biomass of the fish when using the FM and FO used at the 100% SSBP FM and FO inclusion rate. As already mentioned, the mycobacteriosis infection and the lower amino acid profile of the SSBP FM might have impacted the changes in biomass of the fish and affected the outcome. The condition factor, K, is indicative of the leanness/fattiness of a fish, which is generally regarded as an indicator of fish health. K values of <1 indicate thin or poorly nourished fish. Values > 1 indicate the fattiness, which may be species dependent [58]. The condition factor in this study had values between 1.53 and 1.85 for all the treatments, indicating that the fish were well nourished in spite of being affected by mycobacteriosis. Barnham and Baxter [59] gave a K value of 1.40 for a good, well-proportioned fish and 1.6 for excellent condition, trophy class fish (for trout and salmon). The condition factor for striped bass after 1 year when grown in marine conditions ranged between 1.16 and 1.35, depending on the strain of the fish. The condition factor further increased and ranged between 1.38–1.70 in the second year of study [60] The survival percent of fish during the study period did not have a significant difference but the treatment with 100% SSBP FM and FO inclusion had the greatest number of mortalities.

This study found that the total mean amino acid level was found have a higher concentration in fillets as compared to whole fish. Glutamic acid was found to be the highest in concentration followed by lysine [61,62] and the least concentration was found in cysteine followed by orthinine. The amino acid profile generally decreased with the increase in SSBP levels. The difference in the amino acid profile of WFM and SSBP FM (Table 10) may have been the crucial reason for these differences. The whole, fish contains other organs and bones, etc., which may have a different amino acid composition and that might have caused a dilution of amino acids when compared to fillets. The percentage of essential amino acids (EAA) to total amino acids (calculated from Table 8) was higher in fillets (43.97–45.65%) as compared to whole fish (40.28% to 44.92%). The highest percentages of EAA in both types of fish fractions was found in treatment with 25% SSBP. Thus, SSBP could be used at 25% replacement of WFM to have a comparable amino acid profile. While analyzing the fatty acid composition of the fish fillet and the whole fish, it was observed that the saturated fats were higher in whole fish as compared to fillets. Whole fish has fat distributed throughout the body and thus has higher levels of saturated fats. The monounsaturated fats also follow the same trend. In case of polyunsaturated fats (PUFA), their concentration was lower in whole fish as compared to the fillets as well as it decreases as the SSBP inclusion increases in the feed. This is due to the higher PUFA levels in WFO as compared to SSBP FO, as can be seen in Table 11. There were no statistically significant differences in the PUFA content of fillets and whole fish. So, effectively, SSBP FM and FO can be used to replace the traditional FM and FO in fish feeds. The omega-3 and omega-6 content in fillets and whole fish follow opposite trends. While omega-3 decreases as the SSBP content increases, omega-6 increases with increases in SSBP in feeds. This is due to the lower omega-3 and higher omega-6 concentration in SSBP FO as compared to WFO (Table 7). The use of vegetable oils as a replacement for fish oil in commercial salmon feeds is the reason for the higher omega-6 concentration in the fatty acid profile of the farmed fish [63].

During the shelf-life and storage studies, the main reactions occurring in oils and fats that lead to its rancidity are hydrolysis (FFA content) and oxidation (PV and p-Anisidine) [64]. However, the primary measurement of oxidation in omega-3 oils is PV and p-Anisidine [65]. FFA content of oils or oil containing products is an important characteristic of the quality and freshness of oil. The SSBP FM and FO were subjected to accelerated shelf-life conditions of 40 °C and 75% RH for 12 weeks. These conditions are designed to mimic changes in one week that are equivalent to one month of normal ambient condition storage [66]. Thus, 12 weeks of accelerated shelf-life conditions would simulate 12 months of product life under real conditions. It should be noted that the extrapolation of results to real storage conditions would require formal kinetic modeling. The initial lipid oxidation values of FO and FM, and the progression of oxidation during storage, comprise both processing history and quality of the starting material. At the start of the study, the FM showed a FFA content of 3.7 ± 0.06%, whereas FO showed a FFA content of 1.60 ± 0.00%. The FO used in the study was crude, i.e., it did not undergo any refining process other than a reduction in initial water activity. Generally, during the refining stage of oil production, it is either treated chemically (soda neutralization) or physically (deacidification by distillation under high vacuum with steam) to remove the FFA so as to improve the shelf life [67]. One of the reasons this is done is that the process of manufacturing FM and FO exposes the ingredients to higher temperatures and humidity, increasing the risk of oxidation. The FFA is generated due to the hydrolysis/breakdown in ester linkage connecting the fatty acids to the glycerol of the triacylglycerides in the oil in the presence of heat and moisture [68,69]. de Koning et al. [70] suggested that the FFA content of FM is not a reliable quality index because the FFA generated enzymatically increases to a maximum and then decreases. Bimbo [71] suggested that the FFA standard for crude fish oil is about 1 to 7%, but typically ranges between 2 and 5%. While the FFA content for both SSBP FM and FO increased over time, they were within the limits suggested for crude fish oil. The PV is an indicator of primary oxidation (lipid peroxides formed from the reaction of oxygen with the double bond of a fatty acid) and p-Anisidine is a measure of secondary oxidation (aldehydes/ketones formed from the reaction of oxygen with the fatty acid chains in the oil). The FO had higher PV values than FM and vice versa for p-Anisidine values. The FO had a higher PV than FM because FO has more fat content and thus a greater concentration of unsaturated fatty acids, especially the polyunsaturated fatty acids including EPA and DHA, which oxidize easily, even at room temperatures [72]. The peroxide formation increases rapidly during the initial stages of rancidification but decreases over time. Therefore, analyzing p-Anisidine value helps obtain a complete picture as p-Anisidine reacts with the secondary products of oxidation, ketones and aldehydes. Both FM and FO had a decreasing PV value over the 12 weeks of study, indicating that both products underwent peak primary oxidation and were on the decreasing side of the PV graph. As the peroxide values produced the secondary oxidation products, the p-Anisidine value had an incrementally increasing trend over time. The standards adopted by Codex Alimentarius Commission [35] state the standards for fish oil to have a PV of ≤5 meq/kg of oil, p-Anisidine value of ≤20 and TOTOX value of ≤26. Based on these numbers, the FM and FO was not suited for human consumption as the values obtained during the accelerated shelf-life study were higher. Probably, refining process and the addition of antioxidants like ethoxyquin, butylated hydroxytoluene (BHT), butylated hydroxyanisole (BHA), and tocopherols during the processing stage would have helped lower the oxidation process. Additionally, the high temperature and relative humidity to which these products were subjected could have led to a faster deterioration of the quality of these products. Another possible reason for such values of PV and p-Anisidine value could be due to long frozen storage to which the fish were exposed during its entire product life cycle. Tenyang et al. [73] reported that during frozen storage at −18 °C for up to 9 months, the FFA and PV increased, respectively, from 1.35% to 8.06% in oleic acid and 3.77 to 18.62 meq O2/kg in lipid, for red carp (Cyprinus carpio). During frozen storage, the lipids are subjected to hydrolytic and autooxidative changes (which was more evident after 1 month of storage) that are dependent on the degree of fatty acid unsaturation, oxygen exposure, and storage time and temperature [73,74]. The hydrolysis of triglycerides and phospholipids during frozen storage occurs due to the lysis of the cell membranes and the increase in some endogenous enzymes’ activities like hydrolytic enzymes or hydrolases [75]. Several other studies with different fish species have shown an increase in FFA and PV during the low-temperature (frozen) storage of fish [76,77,78,79,80]. Additionally, the raw materials are subjected to heat during the smoking operation, and that may have also contributed towards oxidation of fats in the skins and trimmings. Further, the elevated TOTOX values found for the FO and FM, initially and throughout the shelf-life evaluation, could potentially impact fish health and contribute to the large standard variations in weight and feed intake of the striped bass as shown in Table 4 and Table 5. Fish that consume oxidized fish feed can exhibit a high variation in size and general performance. This outcome stems from factors such as compromised nutritional content and palatability/intake of the feed, oxidative stress and immune system effects. Such consequences of oxidation may have made some of the fish in our study more susceptible to Mycobacterium outbreak [81]. Once the bacteria became established within the system, all groups of fish including the one fed commercial feed could be consistently exposed to its effects.

5. Conclusions

This study shows the potential of FM and FO derived from smoked salmon processing by-products as a viable partial replacement for conventional commercial FM and FO in juvenile domesticated striped bass feed. Despite inherent differences in the nutritional profiles of these FM and FO as compared to the commercial FM and FO, the SSBP FM- and FO-based feeds were equally palatable and supported uniform growth across the 20-week trial in RAS conditions. At substitution levels of 25% and 50%, key performance indicators such as weight gain, specific growth rate (SGR), feed conversion ratio (FCR), protein efficiency ratio (PER), and feed intake showed no significant differences compared to control diets using commercial FM and FO, especially at lower levels of substitution of 25% and 50% SSBP. Although a Mycobacterium outbreak affected all treatments and may have impacted growth outcomes, the overall performance still validated the efficacy of SSBP-derived ingredients. Notably, amino acid and fatty acid concentrations in the SSBP products were lower than their commercial counterparts, a trend observed in the fish fillet and whole-body composition. However, these variations did not compromise growth or feed utilization, especially at lower inclusion rates. The accelerated shelf-life study showed that the FFA values were within the suggested standards for crude FO. The PV and p-Anisidine values were higher potentially due to the long-term frozen storage of the raw material before being used for FM and FO production and/or multitude of processing steps involving heat. Using fresh fish by-products or ones that are not subjected to excess storage time/heat may help produce FM and FO with better shelf-life characteristics. This study highlights a scalable, sustainable solution for small fish processors to transform smoked salmon processing waste into value-added, revenue-generating feed ingredients. By reducing waste and creating high-value applications, the integration of SSBP FM and FO can support circular economy principles within aquaculture. Future research should expand on species applicability, alternative feed formulations, and industrial-scale production to fully unlock the potential of this innovative approach.