sustainability Article Black Soldier Fly Larvae Meal as Alternative to Fish Meal for Aquaculture Feed Marianna Oteri 1 , Ambra Rita Di Rosa 1 , Vittorio Lo Presti 1 , Filippo Giarratana 1 , Giovanni Toscano 2 and Biagina Chiofalo 1, * 1 2 *



 Citation: Oteri, M.; Di Rosa, A.R.; Lo Presti, V.; Giarratana, F.; Toscano, G.; Chiofalo, B. Black Soldier Fly Larvae Meal as Alternative to Fish Meal for Aquaculture Feed. Sustainability 2021, 13, 5447. https://doi.org/10.3390/ su13105447 Academic Editors: Massimo Lucarini and Just Tomàs Bayle-Sempere Department of Veterinary Sciences, University of Messina, 98168 Messina, Italy; marianna.oteri@unime.it (M.O.); dirosaa@unime.it (A.R.D.R.); vittorio.lopresti@unime.it (V.L.P.); filippo.giarratana@unime.it (F.G.) Department of Chemical, Biological, Pharmaceutical and Environmental Sciences, University of Messina, 98166 Messina, Italy; giovanni.toscano@unime.it Correspondence: biagina.chiofalo@unime.it; Tel.: +39-0906766833 Abstract: Hermetia illucens meal (HIM) as ingredient in feed represents a way to achieve more sustainable food production. The aim was to characterize the chemical, microbiological and organoleptic characteristics of four diets for Sparus aurata, isoenergetic and isoproteic, containing 0%, 25%, 35% and 50% of HIM in substitution of fish meal (FM). Analyses were carried out using gas chromatography for fatty acids and amino acids, ICP-OES for minerals and liquid chromatography for aflatoxins and following International Organization for Standardization methods for microbial flora. E-sensing analysis of the diets was evaluated using an artificial sensory platform (E-eye, E-nose and E-tongue). The chemical results were submitted to a one-way ANOVA while Principal Component Analysis (PCA) of the e-sensing data was performed. No significant differences were observed for polyunsaturated fatty acids, thrombogenic and peroxidation indices among the diets. The replacement of FM with HIM increased the content of lysine, methionine, isoleucine, leucine, threonine and valine, while phosphorus, calcium and sodium content decreased (p < 0.01) as the percentage of HIM increased. Lead was significantly below the maximum level set by the EU regulation. The diets showed good hygienic and sanitary quality. The artificial senses permitted distinguishing color, odor and taste among the diets. Data allow considering Hermetia illucens as alternative protein source in fish nutrition. Keywords: Hermetia illucens; aquaculture feed; fatty acids; amino acids; minerals; microbiological quality; e-sensing profile Received: 2 April 2021 Accepted: 11 May 2021 Published: 13 May 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). 1. Introduction Among the costs of the aquaculture farms, feeding represents the largest portion (about 60%) of the economic balance; therefore, the economic success of fish production sector is mainly linked to the use of low-cost nutritionally balanced diets [1]. Therefore, the goal of aquaculture nutritionists, as well as fish farmers, is to obtain a very good conversion ratio to cover the cost of feeding. Proteins with a well-balanced presence of essential and non-essential amino acids are the most important nutrients for the maintenance, growth and feed efficiency of fish. They are the nutrients with the highest cost, and, therefore, their inclusion in the fish feed plays an important role in the overall feed costs. Fish meal, characterized by a high protein content, an excellent amino acid profile, a low carbohydrate level and a high digestibility, is considered the most important feeding source in fish nutrition [2]. However, the significant increase in the price of fish meal, the high protein nutritional needs of fish and the alternating availability in fish meal supply, in recent decades, have led to the study of alternative protein sources in aquafeeds [3]. In relation to the increasing demand for protein sources in animal feeding, there is a great interest towards unconventional sources. In recent years, industrial by-products, Sustainability 2021, 13, 5447. https://doi.org/10.3390/su13105447 https://www.mdpi.com/journal/sustainability Sustainability 2021, 13, 5447 2 of 17 co-products, insects, seaweed ingredients [4] and ex-food or former food products have been investigated as alternative ingredients for livestock and aquaculture feeding. Interest in insects as feed ingredient for terrestrial and aquatic animals continues to grow [5–8]. At present, the exploiting of insects as feed ingredients is not in direct competition with food production. The use of insect meal as a substitute of fish meal seems to represent interesting perspectives for limiting the environmental impact of some production system, such as aquaculture, and for contributing to a circular economy and a “zero waste” society [9]. Among the insect species used as unconventional protein source for fish feeding, Hermetia illucens L. is of the most interesting source for its sustainability related to its capacity to convert organic waste material into biomass containing proteins (40–45%) with high biological value [10], fat (30–35%) with fatty acids of nutritional interest and ash (11–15%) with high mineral concentrations and a high Ca/P ratio [11]. In terms of protein quality, Hermetia illucens larvae contain a favorable essential amino acid profile closer to fish meal than that of soybean meal [5]. The mineral profile and fatty acid composition of the Hermetia illucens larvae were found to be influenced by the diet [12–14]. Liland et al. [14] reported that Hermetia illucens larvae are unable to synthesize polyunsaturated fatty acids, and, therefore, the presence of linoleic acid and alpha-linolenic acids, as well as of eicosapentaenoic and docosahexaenoic in the larvae most likely originates from the substrate. Furthermore, these authors observed that the presence of seaweed in the feeding media can enrich the larvae with macro- and microelements, making the insects a good source of minerals [14]. Nevertheless, Weththasinghe et al. [15] observed a linear decrease in protein and lipid digestibility, protein efficiency ratio and lipid retention in extruded diets for Atlantic salmon as the level of dietary Hermetia illucens meal increases. The use of Hermetia illucens larvae meal (HIM) as a component of feed is a way to achieve more sustainable food production. According to current regulation in Europe [16], the use of HIM is mainly permitted in aquaculture. The optimal level of dietary substitution of fish meal for HIM varies considerably across studies, ranging from 25% to 100%, probably in relation to the different quality of larvae meal, fish species and diet formulation. Furthermore, to obtain a higher-quality fish, it is important to provide good quality and pathogen-free feeds to the fish [17]. In this view, the extrusion method destroys undesirable microbial flora, enzymes and anti-nutritional factors [18–20] and improves the nutritional value of final products and the apparent absorption of minerals [21,22]. Fish have strongly developed chemosensory and chemical signaling systems due to their living in an aquatic environment. The olfactory and gustatory systems comprise the main chemosensory pathways [23]. The combination of artificial senses (Electronic Nose, an Electronic Tongue and an Electronic Eye) proves to be a powerful tool to distinguish different organoleptic profiles related to the different chemical compositions of aquaculture feeds [24]. To further characterize the feasibility of HIM as a unconventional protein source in aquafeeds, the aim of this study was to characterize the chemical, microbiological and mineral composition of diets for Spaurus aurata L. containing HIM as a partial replacement for fish meal (FM). The effects of inclusion of HIM on the organoleptic characteristics of the diets were also investigated with the aim of providing detailed sensory information on the HIM-containing fish meal useful for feed industry. 2. Materials and Methods Four experimental diets were formulated to satisfy the nutritional needs of Sparus aurata. The diets were isoenergetic (about 22 MJ/kg gross energy), isonitrogenous (about 43 g/100 g, as fed) and isolipidic (about 19 g/100 g, as fed). A basal diet (HIM0) with fish meal (FM), as exclusive protein source of animal origin, was prepared. FM was partially replaced with defatted Hermetia illucens meal at 25%, 35% and 50% (as fed basis) in the other three diets (HIM25%, HIM35%, HIM50%) corresponding to the inclusion levels of Sustainability 2021, 13, 5447 3 of 17 0%, 7.9%, 11% and 15.7% of HIM, respectively. The other components of the formulas were adapted to obtain diets with the same energetic content. Diets were prepared by SPAROS Lda (Olhao, Portugal); all dietary ingredients were ground, mixed and extruded using die with 4 mm diameter; the oils were added using a vacuum coating technology. The ingredients of the diets and the proximate composition of the diets (HIM0, HIM25%, HIM35% and HIM50%) are reported in Table 1. Table 1. Diet ingredients and proximate composition of the experimental diets. Ingredients, % as fed Fish meal Hermetia illucens meal Soy protein concentrate Wheat gluten Corn gluten Soybean meal 48 Rapeseed meal Wheat meal Whole peas Fish oil Rapeseed oil Vitamin and mineral premix Vitamin C35 Vitamin E50 Antioxidant Sodium propionate MCP, monocalcium phosphate L-Lysine L-Tryptophan DL-Methionine L-Taurine Chemical composition, % as fed Dry matter Crude protein Crude fat Crude fiber Ash NFE * HIM0 HIM25% HIM35% HIM50% 25.00 0 5.00 5.00 5.00 15.00 5.00 17.45 4.00 5.00 10.00 1.00 0.03 0.02 0.30 0.10 1.50 0.30 0.10 0.20 18.75 7.90 5.00 5.00 5.00 15.00 5.00 15.17 4.00 5.00 9.80 1.00 0.03 0.02 0.30 0.10 2.20 0.35 0.03 0.15 0.20 16.25 11.00 5.00 5.00 5.00 15.00 5.00 14.21 4.00 5.00 9.80 1.00 0.03 0.02 0.30 0.10 2.50 0.37 0.04 0.18 0.20 12.50 15.70 5.00 5.00 5.00 15.00 5.00 12.88 4.00 5.00 9.80 1.00 0.03 0.02 0.30 0.10 2.80 0.40 0.05 0.22 0.20 92.33 42.7 18.6 2.3 9.3 19.43 92.78 42.7 18.6 2.2 9.3 19.98 92.90 42.7 18.6 2.2 9.4 20.00 92.64 42.7 18.7 2.1 9.3 19.84 HIM0, fish meal; HIM25%, HIM35% and HIM50%, Hermetia illucens meal at 25%, 35% and 50% substitution rate of fish meal, respectively. * Nitrogen-free extract, NFE (%) = 100 − (%Crude Protein + %Crude fat + %Crude fiber + %Ash). 2.1. Fatty Acid Analysis Triplicate feed samples were analyzed for the fatty acid composition. Each sample (ca. 2.5 g) added with sodium sulfate (1 g) was pounded manually. The lipids were extracted for 6 h with petroleum ether by a Soxtec™ 8000 Extraction system (FOSS, Padua, Italy). The fatty acid methyl esters (FAMEs) were produced from aliquots of lipids. In detail, sulfuric acid–methanol (1:9, v/v) reagent (2 mL) was added to the extracted lipid samples, and they were then heated at 100 ◦ C for 1 h [25]. FAMEs were analyzed by a Trace 1310 chromatograph (Thermo Fisher Scientific, Milan, Italy) equipped with a flame ionization detector (FID) and a fused silica capillary column (30 m × 0.25 mm I.D., 0.25 µm film thickness) (Omegawax 250; Supelco, Bellefonte, PA, USA) maintained at 100 ◦ C for 5 min, from 100 to 240 ◦ C at 4 ◦ C/min and final isotherm of 240 ◦ C (20 min). Injector and detector temperatures were 250 ◦ C. Injection volume and split ratio were 0.5 µL and 1:50, respectively. The carrier gas was helium at a flow rate of 1 mL/min. Data acquisition was carried out by a Chromeleon Software (Thermo Fisher Scientific, Milan, Italy). The identification of individual compounds was carried out by comparing their retention times Sustainability 2021, 13, 5447 4 of 17 with those of standards (mix 37 FAMEs, Supelco, Inc., Bellefonte, PA, USA). The results were expressed as g/100 g of the total fatty acids identified. Nutritional indices were calculated from the identified fatty acids, as proposed by Ulbricht and Southgate [26], for atherogenic (AI) and thrombogenic (TI) indices, while the equation proposed by Santos-Silva et al. [27] was used for the calculation of the hypocholesterolaemic/hypercholesterolaemic ratio (H/H). Indices were determined according to the following formulas: IA = [C12:0 + (4 × C14:0) + C16:0]/(Σn6-PUFA + Σn3-PUFA + ΣMUFA) (1) IT = (C14:0 + C16:0 + C18:0)/[(0.5 × ΣMUFA) + (0.5 × Σn6-PUFA) + (3 × Σn3-PUFA) + (Σn3-PUFA/Σn6-PUFA)] (2) H/H = (C18:1n9 + C18:2n6 + C20:4n6 + C18:3n3 + C20:5n3 + C22:5n3 + C22:6n3)/(C14:0 + C16:0)] (3) Furthermore, the peroxidation index (PI), which expresses a measure of the peroxidation susceptibility and peroxidative lipid damage for a particular phospholipid membrane, was calculated using the formula reported below [28]: PI = (% dienoic × 1) + (% trienoic × 2) + (% tetraenoic × 3) + (% pentaenoic × 4) + (% hexaenoic × 5) (4) 2.2. Amino Acid Analysis Triplicate feed samples were analyzed for the amino acid composition. For the amino acid analysis, protein hydrolysis and derivatization were performed prior to the separation by GC-FID. Each sample (about 0.25 g) was hydrolyzed in 10 mL of a HCl solution (6 M) at 110 ◦ C for 24 h. During the acid hydrolysis, the asparagine and glutamine were converted to aspartic and glutamic acids [29]; therefore, they were calculated as the sum of the aspartic acid plus asparagine and of the glutamic acid plus glutamine. For cysteine analysis, prior to the acid hydrolysis, a preliminary oxidation was performed for the deamination. Each sample was treated with formic acid:hydrogen peroxide (1:20, v/v) reagent (2 mL) for 30 min at room temperature. Then, a hydrolysis step with a HCl solution (6 M) was performed [30,31]. For the tryptophan analysis, each sample was hydrolyzed in 10 mL of a NaOH solution (4 M) at 112 ◦ C for 16 h, and, after the hydrolysis, each sample was cooled and neutralized with acetic acid [32]. For the chromatographic analysis, procedures for purification, pre-column derivatization and qualitative and quantitative analyses of each amino acid were performed using the EZ:Faast Kit (Phenomenex, Torrance, CA, USA). A Trace 1310 chromatograph (Thermofisher, Waltham, MA, USA) was used, with a flame ionization detector (FID) and a ZB-AAA Amino Acid column (10 m × 0.25 mm ID); the oven temperature was programmed from 110 to 320 ◦ C at 32 ◦ C/min, with a final isotherm of 320 ◦ C (1 min). Injector and detector temperatures were 250 and 320 ◦ C, respectively. Injection volume and split ratio were 2.5 µL and 1:15, respectively. 2.3. Aflatoxin Analysis The analysis of aflatoxins was carried out following the EN ISO method [33] and the criteria suggested by European regulation [34]. For the extraction of the aflatoxins, each sample (about 25 g) of the experimental diet was treated with 125 mL of a methanol:water mixture (7:3, v/v) and 5 g of NaCl. The extract was filtered, diluted with water and passed through an immunoaffinity column (Vicam) containing specific antibodies for aflatoxins B1, B2, G1 and G2. The aflatoxins were isolated, purified and concentrated on column and then recovered with methanol. The above procedure was performed in triplicate on each experimental diet. Aflatoxins were measured by a RP-HPLC coupled to a fluorescence detector (RF), using a post-column derivatization (Kobra cell system). Chromatographic separation was performed using a Nexera LC System (Shimadzu, Milan, Italy), equipped with a Luna column C18 250 × 4.6 mm (lenght × i.d.) and 5 µm of particle size (Phenomenex, Torrance, CA, USA). The mobile phase for the isocratic separation was Sustainability 2021, 13, 5447 5 of 17 a mixture of water:acetonitrile:methanol (3:1:1, v/v/v) with 0.35 mL of nitric acid 4 M and 120 mg/L of potassium bromide. The RF wavelength was set at 365 nm for excitation and 435 nm for emission. The injection volume was 50 µL and the amount of aflatoxins was calculated using an external standard method. The LC method for aflatoxin analyses has been validated for the simultaneous chromatographic determination of total (B1, B2, G1 and G2) and B1 aflatoxins. A calibration curve was constructed for each aflatoxin; the linearity was also tested in the range of 0.05–22 µg/kg, providing a correlation coefficient (R2) of ≥0.9996. The Limit Of Detection (LOD, between 0 and 0.05 µg/kg) and the Limit Of Quantification (LOQ, between 0.05 and 0.13 µg/kg) were calculated by the signal-to-noise (S/N) ratio, which should be greater than 3 and 10, respectively, according to the IUPAC criteria. 2.4. Mineral Element Analysis Triplicate feed samples were analyzed for the mineral composition. Each sample was weighed (about 0.5 g) into an acid-prewashed PTFE vessels and 7 mL of HNO3 at 65% and a Rhenium internal standard (1 mL) were added. The mixture was digested with 1 mL of H2 O2 at 30% using a closed-vessel microwave digestion system (Ethos 1, Milestone, Bergamo, Italy). To validate the analytical method, a Standard Reference Material of spinach leaves (SRM, NIST-1570a) obtained from the National Institute of Standards and Technology (Gaithersburg, MD, USA), was digested using the same analytical method described for the feed samples. An Avio200 ICP-OES instrument (Perkin Elmer, Waltham, MA, USA) equipped with a vertical DualView optical system and a S10 autosampler (Perkin Elmer, Waltham, MA, USA) was used to analyze the mineral content. Table 2 shows the recommended analytical lines length used to perform element analyses; the Argon line at 420.069 nm was used as an internal standard. The applied operational conditions are listed in Table 3. Data were processed using a PerkinElmer Syngistix™ for ICP software (Perkin Elmer, Waltham, MA, USA). Table 2. Analytical lines length (nm) utilized for analysis. Element nm Element nm Element nm Cr Cu Fe B 267.716 327.393 238.204 249.677 K Mg Mn Ca 766.490 285.592 257.610 317.933 Se Zn Na Pb 196.026 213.857 589.592 220.353 Table 3. Operational conditions of the ICP-OES. Parameter Radiofrequency power (W) Plasma gas flow (L/min) Auxiliary gas (L/min) Nebulizer gas (L/min) Sample uptake (mL/min) Conditions 1500 9 0.2 0.7 1 The position of the torch was optimized prior to the analytical phase using the optical optimization procedure of Syngistix™ ICP software with Mn analytical line. All the quantitative measurements were made against external calibration curves constructed from a standard solution of 0.05, 0.25 and 1 ppm of Perkin Elmer (Waltham, MA, USA) for ICP analysis. A Milli-Q ultrapure (Merck Millipore, Merck KGaA, Darmstadt, Germany) water system was used to produce water at 1.8 MΩ/cm for the preparation of solutions and to dilute samples as needed. The calibration curves for all elements were established using the calibration blank and the reagent blank, and all of them resulted with correlation coefficients (r2 ) better than 0.999; the Detection Limits (DLs) of this procedure were determined by Sustainability 2021, 13, 5447 6 of 17 analyzing a matrix blank, which consisted of the same reagents and quantities as those used for sample preparation. 2.5. Microbiological Analysis Twenty-five grams of each experimental diet were homogenized with buffered peptone water (Biolife, Milano, Italy) (ratio of 1:9 w/v) by using a stomacher (400 Circulator; International PBI s.p.a., Milano, Italy) for 60 s at 230 rpm. For each sample, the following parameters were evaluated: (i) enumeration of the aerobic colony at 30 ◦ C [35] on Tryptic Glucose Yeast Agar (Biolife, Milano, Italy) plates, incubated at 30 ± 1 ◦ C for 72 h; (ii) Enterobacteriaceae detection [36] and count [37] on Violet Red Bile Glucose Agar (Biolife, Milano, Italy), incubated at 37 ± 1 ◦ C for 24 h; (iii) enumeration of coliforms [38] on Violet Red Bile Agar (Biolife, Milano, Italy) plates, incubated at 30 ± 1 ◦ C for 24 h; (iv) yeasts and moulds count [39] on Dichloran Glycerol Agar (DG18 Biolife, Milano, Italy), incubated at 25 ± 1 ◦ C for 5 days; (v) detection and enumeration of Clostridium spp. [40] on Tryptose Sulfite Cycloserine Agar (Biolife, Milano, Italy), incubated at 37 ± 1 ◦ C for 24 h in anaerobic conditions; and (vi) detection of Salmonella spp. [41] on Chromogenic Salmonella Agar (Biolife, Milano, Italy) and Xylose Lysine Deoxycholate Agar (Biolife, Milano, Italy) incubated both at 37 ± 1 ◦ C for 24 h. The limit of detection (LOD) was 10 CFU/g for the count of aerobic colonies at 30 ◦ C, Enterobacteriaceae, coliforms, Clostridium spp. and 100 CFU/g for the count of yeasts and molds. Further 25 g of each experimental diet, as previously reported, were homogenized with Listeria Fraser Broth Half Concentration (Biolife, Milano, Italy) for the detection of the Listeria monocytogenes [42], incubated at 30 ± 1 ◦ C for 20 h, followed by a passage in Listeria Fraser Broth (Biolife, Milano, Italy) at 37 ± 1 ◦ C for 24 h and spread both on Agar Listeria according to Ottaviani & Agosti (ALOA® ) (Biolife, Milano, Italy) and Listeria Palcam Agar (Biolife, Milano, Italy) both incubated at 37 ± 1 ◦ C for 24–48 h. 2.6. E-Sensing Analysis The feed samples were analyzed using an artificial sensory platform consisting of an E-eye, E-nose and E-tongue. E-eye: The image was acquired with an artificial vision system (Iris visual analyzer 400, Alpha MOS, Toulouse, France) equipped with a high-resolution charge-coupled device (CCD) camera with 16 million colors. Image acquisition was performed using a top illumination and a white tray at the bottom to easily remove the background contribution (threshold selection applied: R 0-145, G 0-121, B 0-109). Each sample was ground, and the powder was placed and flattened on a plastic Petri dish (diameter 92 mm, height 7 mm). For each sample, 15 images on 15 freshly prepared dishes were acquired. Color spectra were calculated by selecting only the contributions greater than 1%. E-nose: Odor analysis was performed by an electronic nose device (FOX 4000, Alpha MOS, Toulouse, France) equipped with 18 MOS (metal-oxide semiconductor) gas sensors and an automatic headspace sampler (HS100). For each sample, 15 replicates were prepared by placing 2 g of freshly ground fish feed into 10 mL headspace sealed vials. All parameters of the instrument are reported in Table 4. Sustainability 2021, 13, 5447 7 of 17 Table 4. E-nose parameters. Acquisition Oven Duration 120 s Period 1 s Time 1080 s Flow of the carrier gas 150 mL/min Agitator Speed 500 rpm On 5 s Of 2 Incubation time 600 s Incubation temperature 40 ◦ C Syringe Flushing time 120 s Temperature 50 ◦ C Fill speed 500 µL/s Injection Volume 500 µL Speed 500 µL/s E-tongue: Artificial taste analysis was performed using a commercially available electronic tongue (Astree, Alpha MOS, Toulouse, France) equipped with a set of seven potentiometric sensors (ANS, PKS, CTS, NMS, CPS, ANS and SCS), an Ag/AgCl reference electrode (Metrohm, Pte Ltd., Singapore), a mechanical stirrer and a 48-position autosampler. Five grams of each sample were ground and placed in 50 mL of deionized water for 15 min and centrifuged at 3000 rpm for 30 min. The solution was filtered and placed in a 25 mL beaker for the analysis. Single sample analysis was repeated 30 times to obtain the most stable sensor response and the last 15 measurements were considered to perform data processing. The signal was acquired every second for 120 s and the average intensity of the last 20 s was measured. Prior to measurement, the sensors were conditioned using one of the samples as a standard. 2.7. Statistical Analysis The chemical data were analyzed by a one-way ANOVA, using the XLSTAT statistical package [43]. The percentage integration of insect meal (HIM0, HIM25%, HIM35% and HIM50%) was used as a fixed effect. Separation of means was assessed by Tukey’s test, and differences were significant if p < 0.05. A Principal Component Analysis (PCA) of the sensory profile data was performed by Alpha Soft V12.4 (Alpha-MOS, Toulouse, France) to evaluate the discrimination ability between the four experimental diets. The effectiveness of discrimination was assessed by evaluating the discrimination index (DI), which gives the quality of discrimination through an indication of the surface between groups. The DI is calculated automatically by the instrument’s software according to the following formula: DI = 100 × [1 − [(Surface (A) + Surface (B) + Surface (C))/(Total Surface)]] (5) The DI reaches a maximum value of 100 when the groups are completely resolved. 3. Results 3.1. Fatty Acid Profile The fatty acid profile of the four experimental diets is shown in Table 5. Saturated fatty acids did not show any significant (p > 0.05) differences, with the exception of the lauric acid (C12:0), which showed significantly (p < 0.01) higher values in the HIM35% and HIM50% diets than those observed in the HIM0 and HIM25% diets and for the palmitic acid (C16:0) which showed a significant (p < 0.05) higher value in the HIM0 diet than those observed in the diets containing Hermetia illucens meal. Monounsaturated fatty acids show similar content among the experimental diets as well as the polyunsaturated fatty acids, of both the omega 3 and omega 6 series. Table 6 shows the fatty acid classes; the sum of saturated, monounsaturated and polyunsaturated fatty acids; and some indices of nutritional interest, namely the atherogenic (AI) and thrombogenic (TI) indices, the peroxidation index (PI) and the hypocholesterolaemic/hypercholesterolaemic ratio. No significant (p > 0.05) difference was observed among the fatty acid classes, with the exception of the saturated fatty acids which show the highest (p < 0.05) level in the control diet (HIM0). The sum of Sustainability 2021, 13, 5447 8 of 17 the eicosapentaenoic (EPA) and docosahexaenoic (DHA) acids was similar among the diets. Similar values were recorded for AI, TI and PI. The H/H ratio showed significantly (p < 0.05) higher values in the HIM35% and HI50% diets than that recorded in the HIM0 diet while the HIM25% diet showed a value of the H/H ratio similar to those of the other diets. Table 5. Fatty acid composition (g/100 g of fatty acid methyl esters) # of the experimental diets. Fatty Acid HIM0 HIM25% HIM35% HIM50% SEM p C10:0 C12:0 C13:0 C14:0 C15:0 C16:0 C16:1 C17:0 C18:0 C18:1n9 C18:1n7 C18:2 n6 C18:3n6 C18:3n3 C20:0 C20:1n9 C20:2n6 C20:3n3 C20:4n6 C20:5n3 C22:0 C22:1n9 C22:2n6 C23:0 C24:0 C22:6n3 0.04 1.12 B 0.02 2.67 0.23 12.04 a 3.10 0.20 2.72 43.49 3.21 14.34 0.11 4.25 0.47 2.12 0.11 0.30 0.04 4.61 0.25 0.34 0.02 0.22 0.59 3.46 0.03 0.52 C 0.02 2.57 0.22 11.43 ab 3.01 0.20 2.79 44.69 3.19 14.28 0.11 4.36 0.44 2.25 0.11 0.29 0.06 4.56 0.23 0.33 0.02 0.21 0.60 3.54 0.05 1.67 A 0.01 2.55 0.20 11.19 b 2.87 0.19 2.64 43.26 3.10 14.92 0.10 4.45 0.43 1.88 0.10 0.30 0.05 4.92 0.27 0.30 0.03 0.25 0.62 3.69 0.05 1.34 AB 0.01 2.54 0.21 11.17 b 2.99 0.19 2.54 43.55 3.11 14.79 0.10 4.50 0.42 2.05 0.10 0.30 0.05 4.92 0.23 0.27 0.02 0.24 0.61 3.73 0.008 0.080 0.004 0.065 0.009 0.106 0.121 0.008 0.100 0.328 0.041 0.196 0.004 0.088 0.022 0.150 0.004 0.012 0.004 0.246 0.017 0.037 0.013 0.015 0.024 0.228 0.391 0.002 0.615 0.552 0.319 0.013 0.629 0.615 0.445 0.116 0.263 0.184 0.410 0.318 0.465 0.469 0.615 0.943 0.138 0.635 0.398 0.609 0.856 0.384 0.887 0.828 HIM0, fish meal; HIM25%, HIM35% and HIM50%, Hermetia illucens meal at 25%, 35% and 50% substitution rate of fish meal, respectively. # The concentration of fatty acid is expressed as g/100 g, considering 100 g the sum of the areas of all FAME identified. Mean values with different letters within the same row are significantly different, A–C at p < 0.01 and a and b at p < 0.05. Table 6. Fatty acid classes, nutritional indices and in the experimental diets. SFA MUFA PUFA n3-PUFA n6-PUFA EPA + DHA AI TI PI H/H HIM0 HIM25% HIM35% HIM50% SEM p 20.54 a 52.25 27.21 12.61 14.60 8.07 0.30 a 0.24 59.50 4.77 b 19.23 b 53.46 27.32 12.75 14.57 8.10 0.28 b 0.23 60.00 5.10 ab 20.06 ab 51.40 28.54 13.35 15.19 8.61 0.29 ab 0.22 62.90 5.19 a 19.53 ab 51.97 28.51 13.45 15.06 8.65 0.29 ab 0.22 63.10 5.21 a 0.192 0.549 0.667 0.562 0.194 0.470 0.004 0.007 2.376 0.071 0.030 0.198 0.420 0.671 0.185 0.736 0.032 0.232 0.633 0.035 HIM0, fish meal; HIM25%, HIM35% and HIM50%, Hermetia illucens meal at 25%, 35% and 50% substitution rate of fish meal, respectively; SFA, saturated fatty acid class; MUFA, monounsaturated fatty acid class; PUFA, polyunsaturated fatty acid class; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; AI, atherogenic Index; TI, thrombogenic Index; PI, peroxidation Index; H/H, hypocholesterolaemic/hypercholesterolaemic ratio. Mean values with different letters a and b within the same row are significantly different at p < 0.05. Sustainability 2021, 13, 5447 9 of 17 3.2. Amino Acid Profile Table 7 reports the amino acid composition of the experimental diets. Twenty amino acids were identified and quantified; ten of these belong to the indispensable amino acids and ten to dispensable ones. Among the indispensable amino acids, six amino acids (isoleucine, leucine, lysine, methionine, threonine and valine) showed significantly (p < 0.01) higher values in the diets containing insect meal, while histidine showed significantly (p < 0.01) higher values in the HIM35% and HIM50% diets than those observed in the HIM0 and HIM25% diets. Arginine, phenylalanine and tryptophan showed the highest (p < 0.01) values in the control diet (HIM0). Among the dispensable amino acids, hydroxylisine, hydroxyproline and tyrosine showed similar values among the experimental diets. Glutamic acid plus glutamine showed significantly (p < 0.01) higher values in all diets containing the Hermetia illucens meal, while proline and serine showed significantly (p < 0.01) higher values in the diets in which FM has been replaced with HIM at 35% and 50%. The levels of aspartic acid plus asparagine and that of glycine were significantly (p < 0.01) higher in the HIM25% and HIM35% diets than those observed in the HIM0 and HIM50% diets. Cysteine showed the highest (p < 0.05) value in the HIM25% diet and alanine the highest (p < 0.01) level in the HIM0 diet. Table 7. Amino acid composition (g/100 g dry matter) of the experimental diets. Indispensable amino acids Arginine Histidine Isoleucine Leucine Lysine Methionine Phenylalanine Threonine Valine Tryptophan Dispensable amino acids Hydroxylysine Alanine Aspartic acid + Asparagine Cysteine Glycine Glutamic acid + Glutamine Proline Hydroxyproline Tyrosine Serine HIM0 HIM25% HIM35% HIM50% SEM p 2.84 A 1.10 B 1.96 B 3.40 bB 4.66 B 0.80 B 3.06 A 1.66 B 1.79 B 0.12 A 2.33 aB 1.40 C 2.42 A 4.13 aA 5.51 A 0.94 A 2.31 C 2.09 A 2.35 A 0.05 B 2.15 bB 1.60 A 2.52 A 4.22 aA 5.59 A 0.94 A 2.84 B 2.13 A 2.40 A 0.05 B 2.21 abB 1.65 A 2.38 A 4.04 bA 5.53 A 0.94 A 2.87 B 1.96 A 2.26 A 0.05 B 0.026 0.023 0.028 0.024 0.084 0.013 0.005 0.038 0.025 0.002 <0.001 <0.001 0.001 <0.0001 0.004 0.003 <0.0001 0.003 <0.001 <0.0001 0.21 1.48 C 2.37 C 0.21 b 1.76 C 1.38 B 1.88 C 0.37 1.22 2.49 B 0.24 2.04 A 3.30 A 0.39 a 2.30 A 1.64 A 2.44 B 0.66 1.39 2.65 B 0.26 2.02 A 3.16 A 0.19 b 2.26 A 1.72 A 2.76 A 0.65 1.54 2.95 A 0.23 1.81 B 2.85 B 0.17 b 2.04 B 1.74 A 2.82 A 0.60 1.57 2.82 A 0.013 0.013 0.037 0.018 0.023 0.023 0.029 0.055 0.071 0.028 0.149 <0.0001 <0.001 0.006 <0.001 0.001 <0.0001 0.059 0.075 0.001 HIM0, fish meal; HIM25%, HIM35% and HIM50%, Hermetia illucens meal at 25%, 35% and 50% substitution rate of fish meal, respectively. Mean values with different letters within the same row are significantly different, A–C at p < 0.01 and a and b at p < 0.05. 3.3. Mineral Element Profile In Table 8, the average values of minerals in the experimental diets are reported as: macroelements (phosphorus, calcium, potassium, magnesium and sodium), whose needs by the body are in large amounts, microelements (copper, zinc, manganese, iron, etc.), whose needs by the body are in small amounts [44] and toxic metal (lead). The amount of phosphorus, calcium and sodium decreased significantly (p < 0.01) by increasing the amount of HIM to replace fish meal. The presence of insect meal over 25% as a substitute for fish meal (HIM35% and HIM50%) resulted in significantly (p < 0.01) lower levels of potassium, although the HIM35% and HIM50% diets showed similar (p > 0.05) values between them. The Ca/P ratio in the four experimental diets was calculated due to the antagonist effect of these two macroelements [45]. Significantly (p < 0.01) higher levels for Sustainability 2021, 13, 5447 10 of 17 the Ca/P ratio were observed in the HIM0 and HIM25% diets than those of the HIM25% and HIM50% diets. Magnesium was significantly (p < 0.01) higher in the HIM25% diet than in HIM0, HIM35% and HIM50% diets, while these diets did not show (p > 0.05) differences among them. Iron, copper and zinc showed significantly (p < 0.01) higher levels in the HIM35% diet than those of the HIM0, HIM25% and HIM50% diets, except the HIM25% diet, which showed a similar content to the HIM50% diet. A significantly (p < 0.01) higher level of manganese was observed in the HIM25% diet than those of the HIM0, HIM35% and HIM50% diets. The increase in the level of HIM to replace fish meal resulted in a significant (p < 0.05) increase in the amount of chromium. Boron was significantly (p < 0.01) higher in the HIM25% and HIM35% diets compared to the other diets; lead showed the highest (p < 0.01) value in the HIM50% diet (Table 8). Table 8. Mineral element profile (mg/kg dry matter) of the experimental diets. Items Macroelements P—Phosphorus Ca—Calcium K—Potassium Mg—Magnesium Na—Sodium Ca/P ratio Microelements Fe—Iron Zn—Zinc Mn—Manganese Cu—Copper Se—Selenium Cr—Chromium B—Boron Toxic metal Pb—Lead HIM0 HIM25% HIM35% HIM50% SEM p 11,452.63 A 19,560.64 A 11,632.62 A 2045.39 B 8081.56 A 1.71 A 10,966.03 B 18,673.81 B 11,832.99 A 7600.68 A 7575.17 B 1.70 A 10,477.43 C 16,710.51 C 10,553.62 B 2041.30 B 5664.86 C 1.60 B 9557.83 D 15,064.18 D 10,262.06 B 1990.07 B 4675.18 D 1.57 B 10.30 172.16 136.03 78.62 90.82 0.017 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0.001 219.84 C 175.25 C 7.13 D 14.08 C 3.90 B 4.43 b 6.49 B 232.19 B 184.76 B 10.48 A 15.20 BC 4.63 A 5.44 ab 7.82 A 250.24 A 192.10 A 9.86 C 16.56 A 3.26 B 5.62 ab 7.72 A 229.66 BC 184.47 B 10.14 D 16.03 AB 2.84 B 6.03 a 6.60 B 2.38 1.18 0.062 0.27 0.35 0.28 0.18 <0.001 <0.0001 <0.0001 0.001 <0.001 0.019 0.001 0.76 B 0.97 B 0.78 B 1.83 A 0.44 0.001 HIM0, fish meal; HIM25%, HIM35% and HIM50%, Hermetia illucens meal at 25%, 35% and 50% substitution rate of fish meal, respectively. Mean values with different letters within the same row are significantly different, A–D at p < 0.01 and a and b at p < 0.05. 3.4. Mycotoxin Profile The amounts of total aflatoxins (B1, B2, G1 and G2) and aflatoxin B1 in the four experimental diets (HIM0, HIM25%, HIM35%, HIM50%) were below the limit of detection (LOD: 0.05 µg/kg) in all experimental diets. 3.5. Microbiological Profile No significant difference on the load of each microbiological parameter was observed in the experimental diets. For all diets, only a few aerobic colonies with a load below 70 CFU/g (HIM0 = 65 CFU/g, HIM25% = 70 CFU/g; HIM35% = 50 CFU/g; HIM50% = 60 CFU/g) was observed. The counts of Enterobacteriaceae, coliforms, Clostridium spp., yeasts and molds were always under the LOD. Enterobacteriaceae, Salmonella spp., L. monocytogenes and Clostridium spp. were not detected. 3.6. E-Sensing Profile Concerning data provided by the artificial sensory platform, the first step was to separately perform PCAs on the data from E-nose and E-tongue sensors and on the colors of the E-eye code. The next step was to look for the most effective way to combine the data provided by the E-eye, E-nose and E-togue in order to improve the discrimination capability. An intermediate fusion level was adopted in this study. The sensor data with the highest discrimination power were chosen; in particular, the data of four E-nose sensors (LY2/G, LY2/AA, P30/1 and T40/1), three E-tongue sensors (AHS, CTS and NMS) and Sustainability 2021, 13, 5447 11 of 17 four colors extracted from the E-eye (codes 1620, 1890, 1891 and 2147) were chosen. These datasets were reduced, due to the different data size, and a new PCA was performed and the Discrimination Index (DI) calculated. The result is shown in Figure 1. Figure 1. Principal component analysis map for fish feed groups (HIM0, blue; HIM25%, red; HIM35%, green; HIM50%, violet) and loading vectors of selected variables. At the top DI = 93. The first two principal components calculated (PC1 and PC2) explain 99.4% of the total variance and show a DI of 93, highlighting a clear difference among the four diets, in relation to the substitution of fish meal with HIM. The combination of sensor responses with the highest discrimination power of the three artificial senses completely separates the four feed groups. Moreover, the loading plot helps identify the variables accountable for clustering of our dataset, by providing a numerical value which represents the contribution of each original variable to the score plot. Along the PC1 axis, from left to right, the groups are separated mainly by color codes 1890, 1891 and 2147, which are the most distinctive for the HIM0 group. On PC1, the volatile component of the different groups is also distinguished. Figure 1 shows the contribution of LY2/G, LY2/AA, P301 and T401. These sensors show a clear separation between HIM35% and HIM50% diets and the HIM0 and HIM25% diets. Regarding the taste profile, the three selected sensors (AHS, CTS and NMS) help to improve discrimination between groups. In particular, the AHS and CTS sensors have an important contribution on the HIM0 and HIM25% diets, while the NMS sensor mainly distinguishes the groups with higher insect meal content (HIM35% and HIM50%). 4. Discussion Regarding the origin of fatty acids in the larvae, some studies report the possibility of modifying the fatty acid profile through the diet, while others the possibility of an endogenous synthesis of fatty acids. Knowledge of the fatty acid pathway is of great importance if HIM is used as an ingredient in animal feed. Our results are similar to the fatty acid profile determined by Belghit et al. [9] in four experimental diets formulated for Atlantic salmon with an increasing substitution of fish meal with HIM. St-Hilaire et al. [13] observed a reduction of alpha-linolenic acid (C18:3n3), eicosapentaenoic acid Sustainability 2021, 13, 5447 12 of 17 (C20:5n3) and docosahexaenoic acid (C22:6n3) in fish fed a diet containing HIM. This could represent a problem for producers and consumers of the seafood supply chain [46]. However, our results show similar content of the alpha-linolenic acid, eicosapentaenoic acid and docosahexaenoic acid among the diets, despite the increasing addition of HIM in the feed. The content of these fatty acids, essential for the growth and development in fish [47] and associated with disease prevention and health promotion for humans, thanks to the production of anti-inflammatory eicosanoids [48], could be associated with the use, in feeding of the larvae, of a substrate containing fish offal and algae [12,14], since Hermetia illucens is not able to synthesize PUFA [14]. In fact, only plants and marine algae possess the enzymes necessary for the synthesis of linoleic acid (C18: 2n6) and alpha-linolenic acid (the precursor of EPA and DHA) [49]; therefore, similar concentrations of n6-PUFA (mainly linoleic acid) and n3-PUFA among the four diets confirm our hypothesis on the influence of the feeding substrate of Hermetia illucens larvae [46]. Another interesting result concerned the similar values recorded for the health lipid indices (atherogenic and thrombogenic indices), the peroxidation index and the H/H ratio. The atherogenic and thrombogenic indices (AI and TI) take into account, in the formulas used for their calculation, the contribution that each fatty acid has on human health and, in particular, on the probability of influencing the incidence of cardiovascular diseases [26]. The decrease in PI, e.g., during feed storage, indicates the oxidative degradation of PUFA in primary and secondary oxidation products. This process results in a loss of shelf-life, nutritional value and feed safety, as well as reduced consumer acceptability [50]. Finally, the low H/H ratio observed in diets containing HIM suggests some positive nutritional effects on animals. As regards the indispensable amino acids, the substitution of fish meal with Hermetia illucens increased the lysine (EAA) and methionine content in all the levels of inclusion (HIM25%, HIM35% and HIM50%). In fact, Hermetia illucens meal has high levels of amino acids such as methionine and lysine [30]. Furthermore, isoleucine, leucine, threonine and valine showed higher levels in all experimental diets containing Hermetia illucens meal; this fact is nutritionally interesting, as fish cannot synthesize these amino acids de novo [51]. Among the dispensable amino acids (NEAA), the inclusion of HIM at 35% and 50% levels resulted in an increase in proline and tyrosine (NEAA), as observed by De Marco et al. [30] and by Iaconisi et al. [51]. Minerals are present in low quantities in the feed despite having important metabolic roles [22,44]. Mineral elements in fish are involved in many biochemical processes as enzyme cofactors and activators, in the formation of skeletal muscles, in the transmission of nerve impulses and in the acid–base chemical balance [44,52,53]. The low levels of calcium, potassium and sodium observed in the formulation containing Hermetia illucens meal do not represent a problem in the feeding of fish that can readily obtain these minerals from the surrounding water environment [44,53]. On the contrary, the phosphorus requirement must be satisfied through the diet as the fish cannot readily obtain it from the surrounding aquatic environment, which contains a low amount of it [54,55]. Phosphorus and calcium are generally combined together in the fish body, so an adequate Ca/P ratio in the diet is crucial to ensure healthy bone formation and growth performance of fish [45,52]. In relation to their antagonistic effect [56], the dietary Ca/P ratio has to be taken into consideration, as an excess of Ca or P can cause mineral imbalances that can affect the absorption of other minerals (e.g., an excess of Ca into the diet can reduce the absorption of Zn, Fe and Mn), causing some important consequences for bone development, which can be adversely affected when this ratio increases, causing anomalies in mineral homeostasis and bone mass [57] or environmental contamination (excess P is excreted). In our study, the Ca/P ratio of all diets was close to 1, as recommended for fish [58,59]. As for the microelements, such as zinc, essential for the growth and development in fish, cofactor of many enzymes and necessary for the activity of the antioxidant enzyme superoxide dismutase [44,60]; iron, necessary for the blood and muscle pigments [44] and actively involved in oxidation-reduction reactions [60]; and copper, involved in enzymatic activities and oxygen transportation [60], the highest levels were observed in the diet Sustainability 2021, 13, 5447 13 of 17 containing a quote of 35% of defatted Hermetia illucens meal. This result would seem to highlight that the HIM35% diet is the best formula from a nutritional point of view, considering also that fish cannot obtain adequate quantities of these microelements from water [59,61] and that they must receive them through the diet [55]. The marine environment can convey toxic metals, naturally contained or introduced by various human activities through seafood [62]. Fish, both freshwater and marine water, that are at the top of the food chain, are extremely sensitive to exposure to lead (Pb), which is a highly toxic metal [63]. Its toxicity depends on various factors such diet and environment [63]. It seems that Sparus aurata is much more sensitive to waterborne lead exposure than in the diet [64]. However, considering the toxicity of lead, the European regulation [65] has set a maximum level of lead for complete animal feed of 5 ppm, which is considerably higher than the values found in our experimental diets. Fish feeds become rapidly colonized by environmental microbes. Furthermore, all insects are colonized by microorganism and the insect microbiota is generally different from microorganism in the external environment, including ingested food [66,67]. Usually, the presence in commercial fish feeds of spoilage and pathogen bacteria such as Salmonella spp. and Escherichia coli can be related to poor hygienic storage condition, environmental contamination and problems in the extrusion treatment [68,69]. Our results on the microbiological profile of all diets confirm the effectiveness of the extrusion treatment in reducing the microbial load. The low charge observed for aerobic bacteria could be related to a secondary contamination after extrusion treatment. As recent studies have shown that aflatoxin contamination of animal feeds is a frequent issue, the European regulation has set maximum residue limits for total aflatoxins (B1, B2, G1 and G2) and aflatoxin B1 in animal feed [70]. To confirm the effectiveness of extrusion treatment and the good environmental condition during the storage, the total aflatoxins (B1, B2, G1 and G2) and aflatoxin B1 were found to be below the limit of detection in all experimental diets [71]. This is of particular interest because, in aquaculture, their presence in the diet can destroy the availability of certain nutrients, such as vitamin C and thiamine, reducing the immune defense of fish [72]. In addition, the contamination by mycotoxins of fish feed has been reported to cause intoxications [69], tissue abnormalities or liver injury, liver tumor, decreased growth rate and appetite [73] with a decrease of production efficiency and weight of the caught product and an increase in medical costs [74]. The E-nose and E-tongue sensors are non-specific and partially cross-sensitive; therefore, the value of each sensor is not directly related to each other. To better understand the data obtained by the artificial senses and improve the discrimination capability, it is necessary to optimize the analysis through a fusion process often categorized in a three-level model which includes the distinction of low, intermediate and high fusion levels [75,76]. In our study, the use of an intermediate fusion level helped improve the discrimination power. First, some relevant features were extracted from each data source separately, and then they were concatenated into a single array, which was used for multivariate classification and regression [77]. Selected color codes and E-nose and E-tongue sensors identified differences between groups in relation to the percentage of HIM integration. The volatile component was mainly distinguished by LY2/G, LY2/AA, P301 and T401, which are sensors primarily sensitive to volatile organic compounds derived from proteins and lipids such as nitrogencontaining compounds, hydrocarbons and aldehydes [78,79]. These sensors showed a clear separation of the HIM35% and HIM50% groups from HIM0 and HIM25%, in relation to the amino acidic content of the diets. Our observations appear interesting as the use of alternative ingredients in fish diets, in relation to their chemical composition and their level of inclusion, can reduce the acceptability of feeds, even if nutritionally balanced. In fact, different chemical substances can influence fish feeding behavior by acting as attractants through smell or taste [80]. Fish, in general, have a well-defined olfactory sensitivity to amino acids [81–85], which help them locate and identify food. Glycine and alanine are potent odorants that can stimulate feeding behavior by increasing food intake [86,87]. Our results show that HIM-integrated feeds had greater amounts of glycine and alanine com- Sustainability 2021, 13, 5447 14 of 17 pared to HIM0 diet; this probably contributed to the olfactory discrimination performed by the E-nose sensors. Regarding the taste profile, the three sensors selected mainly represent umami taste (NMS), sourness (AHS) and saltiness (CTS). The results of PCA show that the HIM0 and HIM25% diets have a more salty and sour taste, probably related to their high sodium content, while the diets containing insect meal showed higher percentages of glutamic and aspartic acids, representative of the umami taste [88], than those of the HIM0 diet. 5. Conclusions Data suggest that inclusion of Hermetia illucens meal positively influenced the hypocholesterolaemic/hypercholesterolaemic ratio and the content of indispensable amino acids and microelements. The microbiological quality of all diets testifies to the good practices of hygiene and sanitation applied during the production processes of fish feeds. The Esensing analysis permitted distinguishing color, odor and taste in the four feed groups. The combination of sensor responses (E-eye, E-nose and E-tongue) proves to be a powerful tool for discriminating different organoleptic profiles linked to different chemical compositions of experimental diets. This study represents a part of a larger investigation aimed at evaluating the suitability of HIM addition in the Sparus aurata diet through the study of the productive performance, the chemical and organoleptic characteristics of the fillets and the possible development of intestinal inflammation. Author Contributions: Conceptualization, B.C.; methodology, A.R.D.R. and B.C.; software, A.R.D.R.; formal analysis, M.O., A.R.D.R., V.L.P., F.G. and G.T.; investigation, B.C.; data curation, A.R.D.R. and B.C.; writing—original draft preparation, M.O., V.L.P., F.G., A.R.D.R. and B.C.; writing—review and editing, V.L.P., A.R.D.R. and B.C.; supervision, B.C.; and funding acquisition, B.C. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by PO FEAMP 2014–2020 mis. 2.47 CUP J46C18000570006, project codex 03/INA/17 Title of the project “FIFA—Feed Insects for Aquaculture”, Scientific Responsible Biagina Chiofalo. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Conflicts of Interest: The authors declare no conflict of interest. References 1. 2. 3. 4. 5. 6. 7. Mohanta, K.N. Plant feed resources of India. In Invited Papers on Application of Solid State Fermentation Technology in Aquaculture; Sogbesan, O.A., Mohanta, K.N., Sahoo, P.K., Mitra, G., Jayasankar, P., Eds.; Central Institute of Freshwater Aquaculture: Bhubaneswar, India, 2012; pp. 100–113. Zhou, Q.C.; Tan, B.P.; Mai, K.S.; Liu, Y.H. Apparent digestibility of selected feed ingredients for juvenile cobia Rachycentron canadum. Aquaculture 2004, 241, 441–451. [CrossRef] Kumar, V.; Makkar, H.P.S.; Becker, K. Dietary inclusion of detoxified Jatropha curcas kernel meal, effects on growth performance and metabolic efficiency in common carp, Cyprinus carpio (Linnaeus). Fish Physiol. Biochem. 2010, 36, 1159–1170. [CrossRef] Spranghers, T.; Ottoboni, M.; Klootwijk, C.; Ovyn, A.; Deboosere, S.; De Meulenaer, B.; Michiels, J.; Eeckhout, M.; De Clercq, P.; De Smet, S. Nutritional composition of black soldier fly (Hermetia illucens) prepupae reared on different organic waste substrates. J. Sci. Food Agric. 2017, 97, 2594–2600. [CrossRef] [PubMed] Barroso, F.G.; de Haro, C.; Sánchez-Muros, M.J.; Venegas, E.; Martínez-Sánchez, A.; Pérez Bañón, C. The potential of various insect species for use as food for fish. Aquaculture 2014, 422, 193–201. [CrossRef] Makkar, H.P.; Tran, G.; Heuzé, V.; Ankers, P. State-of-the-art on use of insects as animal feed. Anim. Feed Sci. Technol. 2014, 197, 1–33. [CrossRef] Henry, M.; Gasco, L.; Piccolo, G.; Fountoulaki, E. Review on the use of insects in the diet of farmed fish: Past and future. Anim. Feed Sci. Technol. 2015, 203, 1–22. [CrossRef] Sustainability 2021, 13, 5447 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 15 of 17 Rawski, M.; Mazurkiewicz, J.; Kierończyk, B.; Józefiak, D. Black Soldier Fly Full-Fat Larvae Meal as an Alternative to Fish Meal and Fish Oil in Siberian Sturgeon Nutrition: The Effects on Physical Properties of the Feed, Animal Growth Performance, and Feed Acceptance and Utilization. Animals 2020, 10, 2119. [CrossRef] [PubMed] Belghit, I.; Liland, N.S.; Gjesdal, P.; Biancarosa, I.; Menchetti, E.; Li, Y.; Waagbø, R.; Krogdahl, A.; Lock, E.-J. Black soldier fly larvae meal can replace fish meal in diets of sea-water phase Atlantic salmon (Salmo salar). Aquaculture 2019, 503, 609–619. [CrossRef] Sheppard, C.; Newton, G.L.; Burtle, G. Black Soldier Fly Prepupae: A Compelling Alternative to Fish Meal and Fish Oil; A Public Comment Prepared in Response to a Request by the National Marine Fisheries Service; University of Georgia: Tifton, GA, USA, 2007. Yu, G.H.; Chen, Y.H. Research progression on the larvae and prepupae of black soldier fly Hermetia illucens used as animal feedstuff. Chin. Bull. Entomol. 2009, 46, 41–45. St-Hilaire, S.; Cranfill, K.; McGuire, M.A.; Mosley, E.E.; Tomberlin, J.K.; Newton, L.; Sealey, W.; Sheppard, C.; Irving, S. Fish Offal Recycling by the Black Soldier Fly Produces a Foodstuff High in Omega-3 Fatty Acids. J. World Aquac. Soc. 2007, 38, 309–313. [CrossRef] St-Hilaire, S.; Sheppard, C.; Tomberlin, J.; Irving, S.; Newton, L.; McGuire, M.; Mosley, E.; Hardy, R.; Sealey, W. Fly Prepupae as a Feedstuff for Rainbow Trout, Oncorhynchus mykiss. J. World Aquac. Soc. 2007, 38, 59–67. [CrossRef] Liland, N.S.; Biancarosa, I.; Araujo, P.; Biemans, D.; Bruckner, C.G.; Waagbø, R.; Torstensen, B.E.; Lock, E.-J. Modulation of nutrient composition of black soldier fly (Hermetia illucens) larvae by feeding seaweed-enriched media. PLoS ONE 2017, 12, e0183188. Weththasinghe, P.; Hansen, J.Ø.; Nøkland, D.; Lagos, L.; Rawski, M.; Øverland, M. Full-fat black soldier fly larvae (Hermetia illucens) meal and paste in extruded diets for Atlantic salmon (Salmo salar): Effect on physical pellet quality, nutrient digestibility, nutrient utilization and growth performances. Aquaculture 2021, 530, 735785. [CrossRef] European Commission. Commission Regulation (EU) 2017/893 amending Annexes I and IV to Regulation (EC) No. 999/2001 of the European Parliament and of the Council and Annexes X, XIV and XV to Commission Regulation (EU) No 142/2011 as regards the provisions on processed animal protein. Off. J. Eur. Union 2017, 138, 92–116. Olajuyigbe, O.O.; Akande, G.R.; Ezekiel, M.O.; Olusola, A.O.; Salaudeen, M.M.; Amusan, E.E.; Babalola, A.F. Microbial and Proximate Composition of Some Fish Meal Samples. Internet J. Food Saf. 2011, 13, 41–44. Nikmaram, N.; Kamani, M.H.; Ghalavand, R. The effects of extrusion cooking on anti-nutritional factors, chemical properties and contaminating micro-organisms in food. Int. J. Farm. Allied Sci. 2015, 4, 352–354. Alam, M.S.; Kaur, J.; Khaira, H.; Gupta, K. Extrusion and extruded products: Changes in quality attributes as affected by extrusion process parameters: A review. Crit. Rev. Food Sci. Nutr. 2016, 56, 445–473. [CrossRef] Filipovic, S.; Kormanjos, S.; Sakac, M.; Filipovic, J.; Psodorov, D.; Okanovic, D. Effect of extrusion on nutritive value of animal feed. In Proceedings of the 2nd Workshop Feed-to-Food FP7 REGPOT-3. Extrusion Technology in Feed and Food Processing, Thematic Proceedings, Novi Sad, Serbia, 19–21 October 2010; Levic, J., Ðuragić, O., Sredanović, S., Eds.; Institute for Food Technology: Novi Sad, Serbia, 2010; pp. 97–116, ISBN 9788679940223. Razzaq, M.R.; Anjum, F.M.; Khan, M.I. Effect of extruder variables on chemical characteristics of maize (Zea mays L.) extrudates. Pak. J. Food Sci. 2012, 22, 108–116. Alonso, R.; Rubio, L.A.; Muzquiz, M.; Marzo, F. The effect of extrusion cooking on mineral bioavailability in pea and kidney bean seed meals. Anim. Feed Sci. Technol. 2001, 94, 1–13. [CrossRef] Hara, T.J. Olfaction and gustation in fish: An overview. Acta Physiol. Scand. 1994, 152, 207–217. [CrossRef] Zaukuu, J.L.Z.; Bazar, G.; Gillay, Z.; Kovacs, Z. Emerging trends of advanced sensor based instruments for meat, poultry and fish quality—A review. Crit. Rev. Food Sci. Nutr. 2020, 60, 3443–3460. [CrossRef] Christie, W.W. Preparation of Ester Derivatives of Fatty Acids for Chromatographic Analysis. In Advances in Lipid Methodology— Two; Christie, W.W., Ed.; Oily Press: Dundee, Scotland, 1993; pp. 69–111. Ulbritch, T.L.; Southgate, D.A. Coronary Heart Disease: Seven Dietary Factors. Lancet 1991, 338, 985–992. [CrossRef] Santos-Silva, J.; Bessa, R.J.B.; Santos-Silva, F. Effect of genotype, feeding system and slaughter weight on the quality of light lambs. II. Fatty acid composition of meat. Livest. Prod. Sci. 2002, 77, 187–192. [CrossRef] Luciano, G.; Pauselli, M.; Servili, M.; Mourvaki, E.; Serra, A.; Monahan, F.J.; Lanza, M.; Priolo, A.; Zinnai, A.; Mele, M. Dietary olive cake reduces the oxidation of lipids, including cholesterol, in lamb meat enriched in polyunsaturated fatty acids. Meat Sci. 2013, 93, 703–714. [CrossRef] Mustăt, ea, G.; Ungureanu, E.L.; Iorga, E. Protein acidic hydrolysis for amino acids analysis in food—Progress over time: A short review. J. Hyg. Eng. Des. 2019, 26, 81–87, ISSN 1857-8489. De Marco, M.; Martínez, S.; Hernandez, F.; Madrid, J.; Gai, F.; Rotolo, L.; Belforti, M.; Bergero, D.; Katz, H.; Dabbou, S.; et al. Nutritional value of two insect larval meals (Tenebrio molitor and Hermetia illucens) for broiler chickens: Apparent nutrient digestibility, apparent ileal amino acid digestibility and apparent metabolizable energy. Anim. Feed Sci. Technol. 2015, 209, 211–218. [CrossRef] Szkudzińska, K.; Smutniak, I.; Rubaj, J.; Korol, W.; Bielecka, G. Method validation for determination of amino acids in feed by UPLC. Accredit. Qual. Assur. 2017, 22, 247–252. [CrossRef] Doğan, G.; Osman, E.O. Determination of amino acid and fatty acid composition of goldband goatfish [Upeneus moluccensis (Bleeker, 1855)] fishing from the Gulf of Antalya (Turkey). Int. Aquat. Res. 2017, 9, 313–327. [CrossRef] Sustainability 2021, 13, 5447 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 16 of 17 EN ISO 16050:2011. Foodstuffs—Determination of Aflatoxin B1, and the Total Content of Aflatoxins B1, B2, G1 and G2 in Cereals, Nuts and Derived Products—High-Performance Liquid Chromatographic Method (ISO 16050:2003); European Committee for Standardization, Management Centre: Bruxelles, Belgium, 2011. European Commission. Commission Regulation (EC) No 401/2006 of 23 February 2006 Laying Down the Methods of Sampling and Analysis for the Official Control of the Levels of Mycotoxins in Foodstuffs (Text with EEA Relevance). Off. J. Eur. Union 2006, 70, 12–34. ISO 4833:2003. Microbiology of Food and Animal Feeding Stuffs—Horizontal Method for the Enumeration of Microorganisms—ColonyCount Technique at 30 ◦ C; ISO: Geneva, Switzerland, 2003. ISO 21528-1:2017. Microbiology of the Food Chain—Horizontal Method for the Detection and Enumeration of Enterobacteriaceae—Part 1: Detection of Enterobacteriaceae; ISO: Geneva, Switzerland, 2017. ISO 21528-2:2017. Microbiology of the Food Chain—Horizontal Method for the Detection and Enumeration of Enterobacteriaceae—Part 2: Colony-Count Technique; ISO: Geneva, Switzerland, 2017. ISO 4832:2006. Microbiology of Food and Animal Feeding Stuffs—Horizontal Method for the Enumeration of Coliforms—Colony-Count Technique; ISO: Geneva, Switzerland, 2006. ISO 21527-2:2008. Microbiology of Food and Animal Feeding Stuffs—Horizontal Method for the Enumeration of Yeasts and Moulds—Part 2: Colony Count Technique in Products with Water Activity Less Than or Equal to 0.95; ISO: Geneva, Switzerland, 2008. ISO 15213:2003. Microbiology of Food and Animal Feeding Stuffs—Horizontal Method for the Enumeration of Sulfite-Reducing Bacteria Growing under Anaerobic Conditions; ISO: Geneva, Switzerland, 2003. ISO 6579-1:2017. Microbiology of the Food Chain—Horizontal Method for the Detection, Enumeration and Serotyping of Salmonella—Part 1: Detection of Salmonella spp.; ISO: Geneva, Switzerland, 2017. ISO 11290-1:2017. Microbiology of the Food Chain—Horizontal Method for the Detection and Enumeration of Listeria Monocytogenes and of Listeria spp.—Part 1: Detection Method; ISO: Geneva, Switzerland, 2017. XLSTAT. Data Analysis and Statistical Solution for Microsoft Excel; Addinsoft: Paris, France, 2016. FAO. Aquaculture Feed and Fertilizer Resources Information System; FAO: Rome, Italy, 2017. Available online: http://www.fao.org/ fileadmin/user_upload/affris/docs/North_African_Catfish/English/table_4.htm (accessed on 25 March 2021). FAO. Minerals; FAO: Rome, Italy, 2016. Ewald, N.; Vidakovic, A.; Langeland, M.; Kiessling, A.; Sampels, S.; Lalander, C. Fatty acid composition of black soldier fly larvae (Hermetia illucens)—Possibilities and limitations for modification through diet. Waste Manag. 2020, 102, 40–47. [CrossRef] Tocher, D.R. Omega-3 long-chain polyunsaturated fatty acids and aquaculture in perspective. Aquaculture 2015, 449, 94–107. [CrossRef] WHO. Diet, Nutrition, and the Prevention of Chronic Diseases: Report of a Joint WHO/FAO Expert Consultation; World Health Organization: Geneva, Switzerland, 2003. Tvrzicka, E.; Kremmyda, L.S.; Stankova, B.; Zak, A. Fatty acids as biocompounds: Their role in human metabolism, health and disease–a review. Part 1: Classification, dietary sources and biological functions. Biomed. Pap. 2011, 155, 117–130. [CrossRef] [PubMed] Prisacaru, A.E. Effect of antioxidants on polyunsaturated fatty acids—Review. Acta Sci. Pol. Technol. Aliment. 2016, 15, 121–129. [CrossRef] [PubMed] Iaconisi, V.; Secci, G.; Sabatino, G.; Piccolo, G.; Gasco, L.; Papini, A.M.; Parisi, G. Effect of mealworm (Tenebrio molitor L.) larvae meal on amino acid composition of gilthead sea bream (Sparus aurata L.) and rainbow trout (Oncorhynchus mykiss W.) fillets. Aquaculture 2019, 513, 734403. [CrossRef] Braga, W.F.; Araújo, J.G.; Martins, G.P.; Oliveira, S.L.; Guimarães, I.G. Dietary total phosphorus supplementation in goldfish diets. Latin Am. J. Aquat. Res. 2016, 44, 129–136. [CrossRef] National Research Council. Nutrient Requirements of Fish; Subcommittee on Fish Nutrition, The National Academies: Washington, DC, USA, 1993; 124p, ISBN 0-309-59629-7. Tessenderlo, G. The Role of Phosphorus in the Nutrition of Salmond Fish; Tessenderlo Group: Brussels, Belgium, 2005. Tang, Q.; Wang, C.; Xie, C.; Jin, J.; Huang, Y. Dietary available phosphorus affected growth performance, body composition, and hepatic antioxidant property of juvenile yellow catfish (Pelteobagrus fulvidraco). Sci. World J. 2012, 2012, 987570. [CrossRef] Chavez-Sanchez, C.; Martinez-Palacios, C.A.; Martinez-Perez, G.; Ross, L.G. Phosphorus and calcium requirements in the diet of the American cichlid Cichlasoma urophthalmus (Gunther). Aquac. Nutr. 2000, 6, 1–10. [CrossRef] Kumar, V.; Sinha, A.K.; Makkar, H.P.S.; De Boeck, G.; Becker, K. Phytate and phytase in fish nutrition. J. Anim. Physiol. Anim. Nutr. 2011, 96, 335–364. [CrossRef] Hossain, M.A.; Yoshimatsu, T. Dietary calcium requirement in fishes. Aquac. Nutr. 2014, 20, 1–11. [CrossRef] NRC (National Research Council). Minerals. In Nutrient Requirements of Fish and Shrimp; The National Academies Press: Washington, DC, USA, 2011; pp. 163–185. Watanabe, T.; Kiron, V.; Satoh, S. Trace minerals in fish nutrition. Aquaculture 1997, 151, 185–207. [CrossRef] Oliva-Teles, A. Nutrition and health of aquaculture fish. J. Fish Dis. 2012, 35, 83–108. [CrossRef] [PubMed] Tamele, I.J.; Loureiro, P.V. Lead, Mercury and Cadmium in Fish and Shellfish from the Indian Ocean and Red Sea (African Countries): Public Health Challenges. J. Mar. Sci. Eng. 2020, 8, 344. [CrossRef] Sustainability 2021, 13, 5447 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 17 of 17 Lee, J.-W.; Choi, H.; Hwang, U.-K.; Kang, J.-C.; Kang, Y.J.; Kim, K.I.; Kim, J.-H. Toxic effects of lead exposure on bioaccumulation, oxidative stress, neurotoxicity, and immune responses in fish: A review. Environ. Toxicol. Pharmacol. 2019, 68, 101–108. [CrossRef] Souid, G.; Souayed, N.; Yaktiti, F.; Maarou, K. Lead accumulation pattern and molecular biomarkers of oxidative stress in seabream (Sparus aurata) under short-term metal treatment. Drug Chem. Toxicol. 2015, 38, 98–105. [CrossRef] European Commission. Commission Regulation (EU) 2015/186 of 6 February 2015 amending Annex I to Directive 2002/32/EC of the European Parliament and of the Council as regards maximum levels for arsenic, fluorine, lead, mercury, endosulfan and Ambrosia seeds (Text with EEA relevance). Off. J. Eur. Union 2015, 31, 11–17. van der Fels-Klerx, H.J.; Camenzuli, L.; Belluco, S.; Meijer, N.; Ricci, A. Food Safety Issues Related to Uses of Insects for Feeds and Foods. Compr. Rev. Food Sci. Food Saf. 2018, 17, 1172–1183. [CrossRef] Jensen, A.N.; Hansen, S.H.; Baggesen, D.L. Salmonella Typhimurium Level in Mealworms (Tenebrio molitor) After Exposure to Contaminated Substrate. Front. Microbiol. 2020, 11, 1613. [CrossRef] Ubeibi, C.O. Isolation and identification of bacterial isolates from poultry fish feeds sold in Arabica, Delta State, Nigeria. J. Ind. Technol. 2017, 2, 14–20. Uddin, M.A.; Akter, S.; Mimi, M.M. Microbiological analysis of fish feeds with the demonstration of the antibiotic susceptibility of the isolates and the antibacterial activity of the feeds IOSR. J. Pharm. Biol. Sci. 2019, 14, 27–31. European Commission. Commission Regulation (EU) No 574/2011 of 16 June 2011 amending Annex I to Directive 2002/32/EC of the European Parliament and of the Council as regards maximum levels for nitrite, melamine, Ambrosia spp. and carry-over of certain coccidiostats and histomonostats and consolidating Annexes I and II thereto (Text with EEA relevance). Off. J. Eur. Union 2011, 159, 7–24. Nwabueze, A.A.; Nwabueze, E.O. Microbial Flora of Fish Feeds Sold in Asaba, Southern Nigeria. Am. J. Exp. Agric. 2011, 1, 27–32. [CrossRef] Herera, A.A. Histology of Tilapia (Oreochromis niloticus); Bureau of Fisheries and Aquatic Resources: Quezon City, Philippines, 1996; pp. 23–78. Ashley, L.M. Pathology of fish feed Aflatoxins and other anti-metabolites. In A Symposium on Diseases of Fishes and Shell Fishes; Sniesko, S.F., Ed.; American Fisheries Society: Washington, DC, USA, 1970; Volume 5, pp. 366–379. Tacon, A.G.J. Nutritional Fish Pathology. Morphological Signs of Nutrient Deficiency and Toxicity in Farmed Fish; FAO Fish Technical Paper No. 330; FAO: Rome, Italy, 1992; 75p, ISBN 92-5-103267-X. Banerjee, R.; Tudu, B.; Bandyopadhyay, R.; Bhattacharyya, N. A review on combined odor and taste sensor systems. J. Food Eng. 2016, 190, 10–21. [CrossRef] Di Rosa, A.R.; Leone, F.; Cheli, F.; Chiofalo, V. Fusion of electronic nose, electronic tongue and computer vision for animal source food authentication and quality assessment—A review. J. Food Eng. 2017, 210, 62–75. [CrossRef] Borras, E.; Ferre, J.; Boque, R.; Mestres, M.; Acena, L.; Busto, O. Data fusion methodologies for food and beverage authentication and quality assessment—A review. Anal. Chim. Acta 2015, 891, 1–14. [CrossRef] [PubMed] Di Rosa, A.R.; Leone, F. Application of electronic nose systems on animal source food: An overview. In Electronic Nose Technologies and Advances in Machine Olfaction; Albastaki, Y., Albalooshi, F., Eds.; IGI Global: Hershey, PA, USA, 2018; Chapter 8; pp. 151–174. Chiofalo, B.; Di Rosa, A.R.; Lo Presti, V.; Chiofalo, V.; Liotta, L. Effect of supplementation of herd diet with olive cake on the composition profile of milk and on the composition, quality and sensory profile of cheeses made therefrom. Animals 2020, 10, 977. [CrossRef] Hara, T.J. The diversity of chemical stimulation in fish olfaction and gustation. Rev. Fish Biol. Fish. 1994, 4, 1–35. [CrossRef] Goh, Y.; Tamura, T. Olfactory and gustatory responses to amino acids in two marine teleosts—Red sea bream and mullet. Comp. Biochem. Physiol. C Toxicol. Pharmacol. 1980, 66, 217–224. [CrossRef] Hubbard, P.C.; Barata, E.N.; Canário, A.V. Olfactory sensitivity of the gilthead seabream (Sparus auratus L.) to conspecific body fluids. J. Chem. Ecol. 2003, 29, 2481–2498. [CrossRef] [PubMed] Silver, W.L.; Caprio, J.; Blackwell, J.F.; Tucker, D. The underwater electro-olfactogram: A tool for the study of the sense of smell of marine fishes. Experientia 1976, 32, 1216–1217. [CrossRef] [PubMed] Velez, Z.; Hubbard, P.C.; Barata, E.N.; Canário, A.V. Evidence for functional asymmetry in the olfactory system of the Senegalese sole (Solea senegalensis). Physiol. Biochem. Zool. 2005, 78, 756–765. [CrossRef] Yacoob, S.Y.; Browman, H.I.; Jensen, P.A. Electroencephalogram recordings from the olfactory bulb of juvenile (0 year) Atlantic cod in response to amino acids. J. Fish Biol. 2004, 65, 1657–1664. [CrossRef] Knutsen, J.A. Feeding behaviour of North Sea turbot (Scophthalmus maximus) and Dover sole (Solea solea) larvae elicited by chemical stimuli. Mar. Biol. 1992, 113, 543–548. [CrossRef] Velez, Z.; Hubbard, P.C.; Hardege, J.D.; Barata, E.N.; Canário, A.V. The contribution of amino acids to the odour of a prey species in the Senegalese sole (Solea senegalensis). Aquaculture 2007, 265, 336–342. [CrossRef] Zhang, Y.; Pan, Z.; Venkitasamy, C.; Ma, H.; Li, Y. Umami taste amino acids produced by hydrolyzing extracted protein from tomato seed meal. LWT Food Sci. Technol. 2015, 62, 1154–1161. [CrossRef]