Sustainable Aquafeed with Marine Periphyton to Reduce Production Costs of Grey Mullet, Mugil cephalus

Abstract

Fish culture sustainability is improved by reducing the fish product content in aquafeeds. Marine periphyton can be used dually for biofiltering mariculture effluent and fish nutrition. The feasibility of periphyton substituting for fishmeal in aquafeed requires evidence. Toward this goal, four iso-protein (42%) and iso-lipid (10%) aquafeeds for grey mullet (Mugil cephalus) were formulated. A control periphyton-free diet consisted of 32% fishmeal, while in the other three diets, periphyton replaced 25%, 50%, or 100% of the fishmeal. Fish were cultured over 104 days with the four diets while measuring their weight gain, specific growth rate, feed conversion ratio, protein uptake, and production value. In a subsequent 35-day trial, and the utilization and digestibility of the feed and dietary amino and fatty acids were measured after feeding fish with the four diets consisting of 1% of Cr₂O₃ as an indigestible marker. The content of protein, lipids, carbohydrates, and ash in the feed and fish was measured using acceptable protocols while fatty acids and amino acids were measured via gas chromatography and high-performance liquid chromatography, respectively. Economic analyses of the reduction in feed and fish production costs were performed using data from the IndexMundi database and research results. Reducing the fishmeal content to 16% did not negatively affect their growth, feed conversion, or protein uptake. The digestibility of dietary protein and lipids was high under the low-fishmeal diets. Eliminating fishmeal from aquafeed inhibited growth, presumably due to a metabolic deficit in the biosynthesis of long-chain polyunsaturated fatty acids under high-salinity conditions. Economically, dietary periphyton reduced the mullet’s aquafeed and production costs, saving USD 0.8 per kg of produced fish with the low-fishmeal (16%) diet. This diet also reduced the fish-in:fish-out ratio to 2.8. The dual use of periphyton improves fish culture sustainability by recirculating nutrients, saving costs, and reducing the fish-in:fish-out ratio.

1. Introduction

The production of aquatic animals for human consumption is increasing rapidly at an annual rate of 5.3% y⁻¹ (2001–2018), surpassing capture fisheries [1]. Consequently, reliance on formulated feed is also rising [2] and may reach a production yield of over 58 million tons by 2025 [3]. Since aquafeed comprises 50–75% of total fish production costs [4], the sustainability of feed ingredients is of significant concern. Fishmeal (FM) and fish oil (FO) are the most nutritious ingredients in aquafeeds, providing cultivated fish with many of the required protein and lipid macro-nutrients [2,5]. FM is well digested and highly palatable, as well as provides fish with an excellent amino acid profile [6,7]. The long-chain polyunsaturated fatty acids (LC-PUFAs) eicosapentaenoic (EPA; C20:5n-3) and docosahexaenoic (DHA; C22:6n-3) are vital in the diet of marine species [8]. LC-PUFAs are found in large quantities in FO and at a lower content in FM. A deficit in essential fatty acids (EFAs) is unlikely in FM-rich diets, irrespective of the dietary lipid source employed [9].

Fisheries are still the primary source of FM and FO for aquaculture, and these ingredients’ prices are expected to increase further [6]. When fish consumption by aquaculture surpasses the produced biomass, the sector can be defined as a fish ‘consumer’ rather than a ‘manufacturer’. This is demonstrated by the index called fish-in:fish-out ratio (FIFO), which, for many cultured species, is greater than 1. Hence, for each 1 kg of fish biomass produced, more than 1 kg of captured fish is required for the feed preparation [10,11]. Diets with plant meals and oils are rapidly being developed for many fish species, aiming at reducing fish products while providing adequate protein and lipids [12,13]. Such diets potentially reduce the FIFO ratio and improve aquafeed production’s cost-effectiveness and sustainability [2,9].

Periphyton is a complex of sessile aquatic biota, primarily microalgae, bacteria, and their associated detritus, developed on a submerged substrate in the photic layer of various aquatic habitats [14]. In fishponds, periphyton can be produced on artificial substrates, such as bamboo poles and plastic nets, through a relatively inexpensive method [15]. Periphyton can also be produced in the effluent stream from fishponds using minimal resources and energy for growth and harvesting [16]. The produced biomass has two uses: it purifies the water and serves as a nutritious alternative to commercial feeds [17,18,19]. This technology was demonstrated extensively in the culture of freshwater species such as carp [17,20], tilapia [18,21], catfish [22], and shrimps [23]. However, knowledge concerning the feasibility of periphyton as a source of macro- and micro-nutrients in aquafeeds is relatively negligible. The harvested biomass contains 20–40% crude protein in its dry weight [17,24,25] and is wasted unless employed for other applications. In a recent study, periphyton from a biofilter for mariculture effluents efficiently substituted for FM in the diet of gilthead seabream (Sparus aurata) [26]. The replacement of 50% of FM in the diet by periphyton did not harm fish survival, growth, or the feed conversion ratio (FCR) [26]. Moreover, the alternative diet reduced the costs of feed and fish production, as well as the FIFO ratio.

The flathead grey mullet (Mugil cephalus) is a promising species in aquaculture. It presents euryhaline and eurythermal features combined with feeding at the lowest trophic level [27]. However, large-scale farming depends on the availability of nutritious aquafeed at a reasonable price [28]. Periphyton in the culture of mullets has been demonstrated previously in inland saline groundwater [29] and brackish water ponds [27]. In these studies, mullets consumed fresh periphyton in culture ponds alongside commercial aquafeeds. However, evaluating the nutritional value of periphyton and its effects on fish performance is relatively complex in such a configuration. The current study is a novel integration of periphyton into aquafeeds for grey mullets as a substitute for fishmeal. The integration of different levels of periphyton into the feed and the corresponding reduction in FM content allowed us to examine whether and to what extent this ingredient can be used as a novel ingredient substituting for FM in aquafeed.

Replacing fish products in aquafeeds is vital for improving aquaculture sustainability by reducing the fish’s effective trophic level and FIFO ratio [11,30]. Plants have been examined extensively in the diets of various species. Still, the availability of resources like land and water, which are already scarce, to produce the required yields of land crops for this purpose is questionable [31,32]. Extractive species like algae and plants display the circular bioeconomic approach in aquaculture when recovering excess nutrients in fishpond effluent and recycling them back into edible biomass to integrate into the diet of cultured fish, shellfish, or other organisms.

2. Materials and Methods

2.1. Materials

Mullets (Mugil cephalus) were spawned in September 2019 and raised at the National Center for Mariculture (NCM) in the Gulf of Aqaba. A total of 445 fingerlings from the same broodstock were used for the different trials.

2.2. Methods

2.2.1. Experimental Diets

A total of 4 diets, 3 with periphyton and one control diet, were formulated and examined as different treatments in this research. All diets were isoproteic (41.5 ± 1.5% crude protein) and isolipid (10 ± 1.1% crude lipid), as recommended for mullets [28]. Diets were prepared on site by mixing the ingredients in each diet (

Table 1. Formulation and composition of the experimental diets (a) and the proximal composition of macronutrients in marine periphyton (b).

a.
Ingredients, % in DW		Diet
		Control	24% FM	16% FM	0% FM
Periphyton		-	17.4	34.8	34.8
Fishmeal		32	24	16	-
Corn gluten		20	20	20	44
Corn oil		3	3	3	3
Cornstarch		14	4	4	3
Soy Protein		9	9	12	12
Vitamin and mineral premix		2	2	2	2
Wheat bran		20	2	8.2	1.2
Proximate composition
Protein		43.01	42.06	39.43	41.79
Lipid		10.07	9.04	8.83	11.31
Ash		8.78	14.79	22.62	18.7
Moisture		4.37	5.88	6.14	7.06
b.
Periphyton proximate composition, % in DW	Protein		Lipid	Carbohydrate	Ash
	31.71		5.36	44.68	18.25

All values are % of dry weight. Vitamin and mineral premix was formulated by Bar-Magen LTD and contained the following ingredients (per kg): Vitamin A (0.75 MIU); Vitamin D3 (0.1 MIU); Vitamin E (0.5 KIU); Vitamin K3 (0.4 g); Vitamin B1 (0.75 g); Vitamin B2 (2 g); Vitamin B5 (2.5 g); Niacin (7.5 g); Vitamin B6 (1 g); Vitamin B12 (1 mg); folic acid (0.5 g); Biotin (25 mg); Vitamin C (7.5 g); Inositol (10 g); BHT (butylated hydroxytoluene) (10 g); Mn (1.75 g); Zn (7.5 g); Fe (1 g); Cu (0.4 g); I (0.15) g; Co (0.05 g); Se (0.015 g); and corn (879.45 g).

) in an electric horizontal helix ribbon mixer and steam pelleting them into 2.4-mm diameter aquafeeds using a laboratory-model pellet mill die (CL-3, LAB.2584.02., CPM Europe, Zaandam, The Netherlands). The control diet contained FM at a level of 32%. In the other 3 diets, FM was reduced to 24% (FM-24 diet) or 16% (FM-16), or it was eliminated (FM-free diet) (Table 1). Periphyton was absent in the control diet, while its level in the other diets was either 17.4% in FM-24 diet or 34.8% in FM-16 and FM-free diets. In order to compensate for the low-protein content in the periphyton biomass (as compared to FM), the amount of corn gluten was doubled in the FM-free diet. The profile of fatty acid composition in the experimental diets is detailed in

Table 2. Fatty acid composition in the experimental diets (a) and amino acids composition (not including cysteine) in crude periphyton biomass and aquafeed pellets, i.e., control diet (b).

a.
Fatty Acid Composition in mg g−1 Lipid	Diet
	Control	24% FM	16% FM	0% FM	p-Value
C14:0	0.58 ± 0.03 a	0.45 ± 0.08 b	0.36 ± 0.03 b	0.21 ± 0.01 c	<0.01
C16:0	3.57 ± 0.26 a	3.18 ± 0.28 ab	2.75 ± 0.18 b	2.66 ± 0.33 b	0.01
C16:1n-7	0.73 ± 0.08 a	0.55 ± 0.01 b	0.39 ± 0.03 c	0.23 ± 0.02 d	<0.01
C16:02	0.30 ± 0.07 a	0.13 ± 0.09 ab	0.04 ± 0.00 b	0.06 ± 0.08 b	<0.01
C16:03	0.07 ± 0.01 a	0.06 ± 0.00 ab	0.04 ± 0.01 bc	0.02 ± 0.01 c	<0.01
C16:04	0.20 ± 0.05	0.19 ± 0.08	0.16 ± 0.05	0.17 ± 0.04	0.79
C18:0	0.71 ± 0.09 a	0.61 ± 0.07 ab	0.48 ± 0.04 b	0.45 ± 0.06 b	<0.01
C18:1n-9 (OA)	4.42 ± 0.18	4.14 ± 0.46	3.74 ± 0.40	4.11 ± 0.79	0.50
C18:1n-7	0.38 ± 0.05 a	0.34 ± 0.01 ab	0.31 ± 0.02 ab	0.27 ± 0.03 b	0.03
C18:2n-6 (LA)	7.20 ± 0.16	6.78 ± 0.75	5.89 ± 0.60	6.83 ± 1.47	0.40
C17:0	4.73 ± 0.67	4.89 ± 1.92	4.47 ± 1.37	4.31 ± 0.96	0.95
C18:3n-6 (GLA)	0.07 ± 0.02	0.06 ± 0.00	0.05 ± 0.01	0.05 ± 0.02	0.28
C18:3n-3 (ALA)	0.24 ± 0.01 a	0.29 ± 0.03 ab	0.32 ± 0.01 bc	0.37 ± 0.04 c	<0.01
C18:4n-3	0.04 ± 0.04	0.03 ± 0.04	0.06 ± 0.02	0.06 ± 0.01	0.59
C20:0	0.12 ± 0.04	0.10 ± 0.00	0.08 ± 0.01	0.08 ± 0.01	0.19
C20:1n-11	0.13 ± 0.04	0.09 ± 0.01	0.07 ± 0.01	0.07 ± 0.02	0.06
C20:1n-9	0.15 ± 0.10	0.13 ± 0.05	0.11 ± 0.04	0.12 ± 0.02	0.87
C20:2n-6	0.12 ± 0.08	0.10 ± 0.01	0.04 ± 0.02	0.04 ± 0.01	0.15
C20:3n-6 (DGLA)	0.10 ± 0.11	0.03 ± 0.04	0.02 ± 0.01	0.01 ± 0.01	0.34
C20:4n-3	0.22 ± 0.08	0.20 ± 0.01	0.16 ± 0.02	0.16 ± 0.04	<0.01
C22:1n-11	0.07 ± 0.01	0.07 ± 0.00	0.06 ± 0.00	0.06 ± 0.01	0.28
C22:1n-9	0.17 ± 0.01 a	0.14 ± 0.02 a	0.08 ± 0.01 b	0.05 ± 0.01 b	<0.01
C22:1n-7	0.16 ± 0.06	0.13 ± 0.03	0.10 ± 0.02	0.10 ± 0.03	0.27
C22:5n-3 (DPA n-3)	0.10 ± 0.03 a	0.08 ± 0.01 ab	0.06 ± 0.01 ab	0.02 ± 0.02 b	0.01
∑ FAs	21.73 ± 1.67 a	19.24 ± 1.75 ab	16.26 ± 1.55 b	16.56 ± 2.68 ab	0.04
∑ SFA	4.98 ± 0.38 a	4.33 ± 0.43 ab	3.67 ± 0.26 b	3.40 ± 0.40 b	<0.01
∑ MUFA	6.21 ± 0.53	5.59 ± 0.55	4.86 ± 0.51	5.01 ± 0.81	0.11
∑ PUFA	10.53 ± 0.76 a	9.32 ± 0.77 ab	7.73 ± 0.79 b	8.16 ± 1.48 ab	0.05
∑ n-3 PUFA	2.35 ± 0.28 a	1.85 ± 0.05 b	1.36 ± 0.11 b	0.89 ± 0.19 c	<0.01
∑ n-6 PUFA	7.62 ± 0.39	7.10 ± 0.74	6.12 ± 0.62	7.02 ± 1.51	0.34
n-3/n-6	0.31 ± 0.02 a	0.26 ± 0.02 ab	0.22 ± 0.00 b	0.13 ± 0.03 c	<0.01
C22:6n-3 (DHA)	0.88 ± 0.12 a	0.62 ± 0.02 b	0.36 ± 0.04 c	0.14 ± 0.04 d	<0.01
C20:5n-3 (EPA)	0.87 ± 0.07 a	0.63 ± 0.02 b	0.40 ± 0.04 c	0.14 ± 0.03 d	<0.01
C20:4n-6 (ARA)	0.08 ± 0.02 a	0.07 ± 0.00 ab	0.05 ± 0.00 bc	0.03 ± 0.00 c	<0.01
C22:5n-6 (DPA n-6)	0.06 ± 0.01	0.06 ± 0.03	0.07 ± 0.02	0.07 ± 0.02	0.87
DHA/EPA	1.01 ± 0.07	0.98 ± 0.00	0.90 ± 0.05	0.99 ± 0.06	0.17
b.
Amino acid composition 1 in µg g−1 protein		Pellets		Periphyton
4-Hydroxyproline		0.51		0.89
Alanine		28.99		28.33
* Arginine		1.53		4.71
Aspartic acid		59.38		38.72
Glutamic acid		147.72		35.33
Glycine		21.21		20.58
* Histidine		12.84		5.77
* Isoleucine		46.64		39.50
* Leucine		91.24		64.80
* Lysine		21.35		27.68
* Methionine		20.93		5.05
* Phenylalanine		46.34		40.24
Proline		75.23		34.26
* Threonine		14.73		13.70
* Tyrosine		18.88		22.07
* Valine		101.51		85.96

OA = oleic acid; LA = linoleic acid; GLA = gamma-linolenic acid; ALA = alpha-linolenic acid; DGLA = Dihomo-γ-linolenic acid; DHA = docosahexaenoic acid; EPA = eicosapentaenoic acid; ARA = arachidonic acid; DPA = Docosapentaenoic acid. SFAs = saturated fatty acids; MUFAs = monounsaturated fatty acids; PUFAs = polyunsaturated fatty acids; LC-PUFAs = long-chain polyunsaturated fatty acids. Significant differences are presented as different superscripts in each row (p < 0.05). Data are presented as mean ± SD of technical replicates per diet (2 for the control and 3 for the 24% FM, 16% FM, and 0% FM diets). Statistical analysis was performed by one-way ANOVA, followed by Tukey’s multiple comparisons test. ¹ Cysteine was not measured. * Essential amino acids.

. The control diet consisted of a significantly higher amount of C14:0 and C16:1n-7. The experimental diets with periphyton and low or no FM (16% or FM-free diets) presented a higher level of ALA (C18:3n-3) (p < 0.05) but lower levels of DPA n-3 (C22:5n-3) and ARA (C20:4n-6), as compared to those of the control diet. Diets with periphyton at a higher inclusion rate (34.8% DW) presented lower contents (p < 0.05) of EPA (C20:5n-3), DHA (C22:6n-3), C16:0, C18:0, C16:2, C16:3, and C22:1n-9, as compared to those of the control diet (p < 0.05). The amino acid composition in crude periphyton was relatively similar to that in the control aquafeed (Table 2), although the level of glutamic acid and methionine was 4-fold higher in the aquafeed, and arginine content was 3-fold higher in the periphyton (Table 2).

2.2.2. Fish Feeding Trial

The 445 fingerlings of Mugil cephalus at an average body weight of 12.8 ± 3.7 g were distributed over 20 cylindrical 370 L polypropylene tanks with a steep conical bottom (with 22 ± 1 fish in each) placed in a semi-enclosed greenhouse. The 20 fish tanks were randomly assigned to the 4 types of diet (including the control), while 5 tanks were assigned for each diet as replicates. The duration of feeding trial was 104 days, in which fish were fed manually three times a day (at 9:00, 12:00, and 15:00) at a feeding rate of 4% of their body weight (BW). High water quality was maintained by pumping in seawater at a high flow rate. For this, seawater from the Gulf of Aqaba was pumped from a depth of 20 m 300 m offshore. The culture tanks were provided with continual aeration using air stones. Daily maintenance included adjusting the seawater flow rate and removing solid waste. Every two weeks, during fish weighing, the culture tanks were drained and cleaned to keep the rearing environment safe from pathogens and in optimal condition. Water temperature, dissolved oxygen, and pH were monitored in the fish tanks daily at 8:00 a.m. using a portable device (OxyGuard© Handy Gamma; Farum, Denmark). Every two weeks, water from fish tanks was sampled and filtered (0.45 μm GF/C, Whatmann, Sigma-Aldrich, Rehovot, Israel) before determining the concentration of total ammonia nitrogen (TAN), nitrite (NO2-N), nitrate (NO3-N), and phosphorus (PO4-P) following recommended protocols for analyses in seawater [33,34]. A similar sampling regime occurred for alkalinity and pH measurements in unfiltered water. Throughout the entire experimental period, the temperature of the culture water ranged between 21.2 and 26.6 °C with no significant differences between fish tanks (p = 0.9811); oxygen level was above 6 mg/L, and the mean values of pH and alkalinity were 8.42 and 2.50 meq/L, respectively (

Table 3. Water quality parameters during the 104-day feeding trial.

	Diet
	Control	24% FM	16% FM	0% FM	p-Value
Temperature, °C	23.90 ± 1.23	23.87 ± 1.23	23.89 ± 1.23	23.90 ± 1.24	0.98
DO, mg/L	6.47 ± 0.34 a	6.52 ± 0.36 b	6.54 ± 0.36 b	6.53 ± 0.37 b	0.02
TAN, mg/L	0.08 ± 0.16	0.03 ± 0.07	0.02 ± 0.04	0.01 ± 0.01	0.07
Nitrite, mg/L	-	-	-	-	-
Nitrate, mg/L	-	-	-	-	-
Phosphate, mg/L	0.04 ± 0.01	-	0.0 ± 0.01	0.03 ± 0.11	0.33
pH	8.42 ± 0.05	8.42 ± 0.04	8.42 ± 0.07	8.42 ± 0.05	0.40
Alkalinity, meq/L	2.51 ± 0.13	2.47 ± 0.13	2.55 ± 0.15	2.47 ± 0.14	0.21

DO = dissolved oxygen; TAN = total ammonia nitrogen. Data are presented as mean ± SD of 5 replicates per dietary treatment. Significant differences are presented as different superscripts in each row (p < 0.05). Statistical analysis was performed by one-way ANOVA, followed by Tukey’s multiple comparisons test.

). Water quality conditions in the different tanks were relatively constant and similar during the trial (Table 3) and within the acceptable range for cultivating marine fish [35]. Maximal levels of TAN and phosphorus were 0.08 mg/L and 0.04 mg/L, respectively, while neither nitrite nor nitrate was detected (Table 3).

2.2.3. Diet and Nutrient Digestibility

At the end of the feeding trial, a continuing trial was performed to measure the apparent digestibility coefficient (ADC) of the four diets. For this, we used the same diets but with a minor modification: the integration of 1% of chromic oxide (Cr₂O₃) into each diet, as an indigestible marker [36]. The experiment took place in the same greenhouse. Nine fish from each diet treatment were transferred into three different cylindrical polypropylene tanks (three fish in each tank) of 33 L with a steep conical bottom. Fresh seawater entered the tank at the top, perpendicular to the tank wall, at a flow rate of about 90 L h⁻¹. After allocation to the new tanks, fish were allowed to acclimate to the new environment over three days while being fed regularly. Following this period, fish were fed over 35 days with the Cr₂O₃-marked diets while each fish had the same diet type as they received in the feeding trial. Every day at 13:00 h, food was distributed manually to the different tanks for 2 h to allow for maximal consumption of the daily quota (4% BW). The uneaten feed was then siphoned to eliminate feces contamination with food particles, and the water circulation was closed to eliminate any leakage of feces. On the following day (9:00 h), feces were collected from the bottom outlet of the tanks through an external vertical standpipe and transferred to a settling cone (1 L). The samples feces were washed with distilled water to eliminate salt residues that may interfere with downstream analyses. Feces were then dried at 60 °C for 22 h and stored at −20 °C pending analyses. At the end of the digestibility trial, fish were fasted for 24 h, sacrificed using MS222, and then kept at −20 °C pending further analyses. During the 35-day period of the experiment, the ambient conditions in the culture tanks were sufficient for fish culture with mean temperature of 25.5 °C, oxygen level of >6 mg/L, and a mean pH value of 8.42.

2.2.4. Fish Performance

All fish were measured individually for their body weight at the start and end of the feeding experiment. In addition, fish biomass in each tank was measured every 2 weeks by weighing the fish in groups of 5 individuals at a time. Fish weighing was performed after sedation with clove oil. Fish mortality and morbidity were documented throughout the experiment. After processing all the data, the following recommended equations [37] were used for calculation of fish performances in terms of specific growth rate (SGR), percentage weight gain (PWG), feed conversion ratio (FCR), protein efficiency ratio (PER), and protein productive value (PPV):

SGR (%BW d⁻¹) = {[(ln (Wf)) − (ln (Wi)]/t} × 100

PWG (%) = ((Wf − Wi)/Wi) × 100

FCR (AFDW g/g) = FG/(Wf − Wi)

PER (g WW/g) = (Wf − Wi)/FP

PPV (%) = (Tp/FP) × 100

where: Wi = initial wet weight biomass per fish tank (g WW); Wf = final wet weight biomass per fish tank (g WW); t = time (days); AFDW = ash-free dry weight; FG = total provided feed (g); FP = total protein given per fish tank (g); and Tp = total protein retained in tissue.

2.2.5. Biochemical Analyses

At the end of the experiment, fish were fasted for 24 h to prevent the undesired effects of feed residues in the gut on analyses of fish biochemical composition. For these analyses, 9 fish from each tank were sampled randomly and sacrificed by bathing them in a separate tank that contained MS222. Out of each set of 9 fish, the whole-body and fatty acid proximal compositions in the fillet were analyzed for 7 and 2 fish, respectively. All other fish were kept in the culture tanks and fed their original diets until the initiation of the digestibility trial. The proximal composition of protein, lipids, ash, and moisture was determined in the aquafeeds, fish tissue, and feces samples. Moisture was measured by thermal drying a known amount of biomass in an oven at 100 °C for 24 h and measuring the weight loss. Ash content was determined by burning the dry biomass in porcelain crucibles in a muffle furnace at 500 °C for 6 h and calculating the ash-free dry weight (AFDW). Following thermal drying for whole-body analysis of fish, samples were cut into pieces and minced in a coffee grinder until a homogenate subsample of each fish was obtained. Nitrogen levels were determined by the Kjeldahl method [38], and protein content was calculated using an N-protein factor of 6.25 [39]. Crude lipid was measured following chloroform–methanol extraction [40]. The analysis of fatty acid (FA) content in the aquafeeds and fish fillets included freeze-drying the samples for 23.75 h in a lyophilizer. FA methyl esters (FAMEs) were measured in a certified laboratory at the Jacob Blaustein Institute for Desert Research (BIDR). FAMEs were measured by gas chromatography (TRACE GC Ultra, Thermo, Riva del Garda, Italy) after their extraction in methanol (with 14% boron trifluoride) and sonication [41]. The programmed temperature vaporizer injector was set to increase the temperature from 40 °C at the time of injection to 300 °C at the time of sample transfer. The separation was achieved on a fused-silica capillary column (30 m × 0.32 mm) using the following oven temperature program: 1 min at 130 °C, followed by a linear gradient to 220 °C, and finally 10 min isocratic elution at 220 °C. Helium was used as the carrier gas at a flow rate of 2.5 mL min⁻¹. The detector temperature was fixed at 280 °C. FAMEs were identified, and their level was measured according to co-chromatography with authentic standards (Sigma-Aldrich, Rehovot, Israel). The content of amino acids in feed pellets and fish tissue was analyzed in a certificate laboratory (Amino-Lab, Nes-Ziona, Israel) using high-performance liquid chromatography (HPLC) following recommended protocols.

2.2.6. Apparent Digestibility Coefficient Analysis

ADC was determined according to the level of Cr₂O₃ in the feed pellets versus that in fish feces. The concentrated Cr₂O₃ in fish feces relative to the digestible matter assesses diet (or nutrient) digestibility [36]. Chromic oxide level in diets and fecal samples was determined by the method of perchloric–sulfuric acid digestion with molybdenum as a catalyst [42]. Approximately 100 mg of feces or 500 mg of feed were weighed into a Kjeldahl digestion flask, and 10 mL of the oxidation mix was added to each sample. The mixture was boiled at 280 °C for about 30 min. After the solution turned from green to orange, it was left to cool, and distilled water was added to the sample. The solution was then transferred to a 50 mL volumetric flask and diluted to volume. The chromic oxide content of the oxidized solution was then measured directly against a known standard by spectrophotometry at 350 nm. The ADC of experimental feeds was estimated following the recommended equation [43]:

ADC (%) = 100 − [100 × (Cr₂O₃ Feed/Cr₂O₃ Feces) × (Nutrient in Feces/Nutrient in Feed)]

2.2.7. Economic Analysis and FIFO Calculation

An economic analysis was performed to evaluate the cost of the experimental diets and fish production. Feed cost was calculated based on the market price of the different ingredients in the formulated feeds. Prices true to the day of diet formulation were taken from the commercial feed production factory Zemach Feed Mill LTD and Index Mundi database [44]. The following equations were used for feed and fish production costs:

Feed cost (USD kg⁻¹ feed) = ∑ ingredient cost (USD kg⁻¹) × ingredient content in feed (%).

Fish production (USD kg⁻¹ produced fish) = Feed cost (USD kg⁻¹ feed) × FCR (as measured in the current feeding trial, excluding dead fish).

The FIFO ratio demonstrates the balance between the biomass of captured fish required for the fish products in the diet and the produced fish biomass when fed this diet [11,12]. We used Jackson’s (2009) [45] estimation for yields of FM and FO from biomass of captured fish, which were 22.5 and 5%, respectively. We calculated the FIFO ratio for grey mullets fed with each of the experimental diets in the current study and for other studies of this fish under different diets. The following equation was used for the FIFO ratio [45]:

FIFO Ratio = [(Level of FM in the diet + Level of FO in the diet)/(Yield of FM from wild fish + Yield of FO from wild fish)] × FCR

2.2.8. Statistical Analysis

Data were analyzed using one-way analysis of variance (ANOVA) and Tukey’s multiple significant difference tests using the GraphPad Prism software program (V8.0.1). Differences were considered significant at p < 0.05. Mean values of SGR, FCR, PER, PPV, and PWR were compared, and averages of the abiotic parameters and ADC were compared. An unpaired t-test was performed to compare fish biochemical composition in the feeding and digestibility trials.

3. Results

3.1. Fish Performances

Among the four examined diets, fish presented the most rapid growth and weight gain when fed the control feed (

Table 4. Growth performance, survival, and protein uptake by Mugil cephalus when fed experimental diets over 104 days.

	Diet
	Control	24% FM	16% FM	0% FM
SGR, % d-1	0.79 ± 0.10 a	0.69 ± 0.06 a	0.64 ± 0.05 ab	0.54 ± 0.05 b
PWG, %	126.29 ± 25.82 a	99.76 ± 17.37 a	80.23 ± 23.47 b	64.91 ± 15.3 b
FCR	3.74 ± 0.54 a	3.89 ± 0.16 a	3.89 ± 0.3 a	5.02 ± 0.4 b
PER	0.58 ± 0.1 a	0.52 ± 0.02 ab	0.51 ± 0.04 ab	0.41 ± 0.03 b
PPV, %	33.31 ± 4.97 a	30.01 ± 3.32 ab	28.69 ± 5.71 ab	20.45 ± 3.26 b
Survival, %	100 a	98.22 ± 2.44 ab	97.35 ± 3.91 ab	90.87 ± 7.15 b

Data are presented as mean ± SD from. Significant differences are presented as different superscript in each row (p < 0.05). Statistical analysis was performed by one-way ANOVA, followed by Tukey’s multiple comparisons test. SGR = specific growth rate; PWG = percentage of weight gain; FCR = seed conversion ratio; PER = protein efficiency ratio; and PPV = protein productivity value. For FCR, PER, and PPV, dead fish were excluded from the calculations.

). A slight decrease in weight gain and SGR was measured when not removing the entire FM from the aquafeed and adding periphyton up to the level of 34.8% (Table 4). However, a significant growth deficit was measured when removing the entire FM content from the feed with a poorer SGR and a weight gain about 50% less than that under the control diet (Table 4). The FCR, PER, and PPV were not harmed significantly when reducing the FM in the aquafeed to 24% or 16% (half of the control feed).

3.2. Fish Body Biochemical Composition

The whole-body proximal composition of protein, lipids, ash, and moisture in the mullets fed the different diets was relatively similar (Table 4). The protein content ranged between 55.21% and 58.47% of their DW and lipids between 25.91% and 31.06% (

Table 5. Whole-body biochemical composition of Mugil cephalus after feeding them experimental diets for 104 days.

Ingredients (% in DW)	Diet
	Control	24% FM	16% FM	0% FM	p-Value
Protein	56.71 ± 4.94	56.5 ± 4.6	58.47 ± 2.86	55.21 ± 7.17	0.60
Lipid	27.98 ± 5.99	28.71 ± 4.18	25.91 ± 4.51	31.06 ± 7.24	0.43
Ash	14.32 ± 2.53	14.22 ± 1.88	16.22 ± 3.79	13.31 ± 3.23	0.12

Data are presented as mean ± SD of 13, 15, 12, and 10 replicates for the control, 24% FM, 16% FM, and 0% FM diets, respectively. Statistical analysis was performed by one-way ANOVA, followed by Tukey’s multiple comparisons test.

). The fish fillet was richer in C18:3n-6 and C20:3n-6 after feeding them the 0% FM diet (p < 0.05) (

Table 6. Effects of fish meal replacement by periphyton on fatty acid composition in fillet of Mugil cephalus after feeding them experimental diets for 104 days.

Fatty Acids, mg/g Lipid	Diet
	Control	24% FM	16% FM	0% FM	p-Value
C14:0	0.61 ± 0.45	0.54 ± 0.33	0.62 ± 0.40	0.85 ± 0.59	0.69
C16:0	4.95 ± 3.72	4.59 ± 2.47	5.49 ± 3.06	7.06 ± 4.57	0.43
C16:1n-7	1.31 ± 1.01	0.89 ± 0.60	0.95 ± 0.67	1.56 ± 1.18	0.54
C16:02	0.32 ± 0.06	0.34 ± 0.11	0.35 ± 0.07	0.36 ± 0.07	0.71
C16:03	0.12 ± 0.05	0.12 ± 0.03	0.14 ± 0.03	0.18 ± 0.19	0.51
C16:04	0.21 ± 0.07	0.19 ± 0.07	0.22 ± 0.06	0.22 ± 0.03	0.65
C18:0	1.02 ± 0.32	1.01 ± 0.27	1.15 ± 0.42	1.14 ± 0.43	0.74
C18:1n-9 (OA)	4.40 ± 4.18	4.42 ± 3.05	4.76 ± 3.38	6.61 ± 4.95	0.55
C18:1n-7	0.56 ± 0.47	0.48 ± 0.27	0.59 ± 0.36	0.75 ± 0.40	0.46
C18:2n-6 (LA)	5.63 ± 4.69	6.39 ± 4.09	7.30 ± 4.76	8.53 ± 5.57	0.57
C17:0	4.93 ± 0.92	5.07 ± 0.37	4.99 ± 0.59	4.66 ± 0.54	0.51
C18:3n-6 (GLA)	0.25 ± 0.15 a	0.28 ± 0.13 a	0.37 ± 0.15 a	0.81 ± 0.51 b	<0.01
C18:3n-3 (ALA)	0.19 ± 0.16	0.25 ± 0.16	0.34 ± 0.28	0.43 ± 0.30	0.13
C18:4n-3	0.20 ± 0.21	0.12 ± 0.04	0.14 ± 0.11	0.16 ± 0.07	0.53
C20:0	0.12 ± 0.10	0.10 ± 0.04	0.10 ± 0.04	0.13 ± 0.09	0.74
C20:1n-11	0.22 ± 0.20	0.22 ± 0.13	0.23 ± 0.12	0.24 ± 0.13	0.99
C20:1n-9	0.28 ± 0.15	0.22 ± 0.04	0.22 ± 0.08	0.29 ± 0.17	0.42
C20:2n-6	0.29 ± 0.16	0.28 ± 0.11	0.36 ± 0.15	0.40 ± 0.18	0.25
C20:3n-6 (DGLA)	0.22 ± 0.15 a	0.17 ± 0.07 a	0.25 ± 0.13 a	0.45 ± 0.20 b	<0.01
C20:3n-3	0.22 ± 0.13	0.18 ± 0.07	0.21 ± 0.10	0.21 ± 0.13	0.86
C20:4n-3	-	-	-	-	-
C22:1n-11	0.09 ± 0.09	0.11 ± 0.05	0.08 ± 0.04	0.08 ± 0.05	0.66
C22:1n-9	0.19 ± 0.13	0.16 ± 0.09	0.15 ± 0.06	0.14 ± 0.05	0.62
C22:5n-3 (DPA n-3)	0.42 ± 0.18	0.39 ± 0.18	0.34 ± 0.11	0.29 ± 0.09	0.22
∑ FAs	24.01 ± 17.53	23.27 ± 12.32	25.94 ± 14.23	32.19 ± 19.34	0.60
∑ SFA	6.71 ± 4.81	6.24 ± 3.19	7.36 ± 3.98	9.18 ± 5.70	0.49
∑ MUFA	7.05 ± 6.60	6.50 ± 4.12	6.98 ± 4.80	9.67 ± 6.84	0.60
∑ PUFA	10.25 ± 6.20	10.53 ± 5.07	11.61 ± 5.78	13.34 ± 6.86	0.66
∑ n-3 PUFA	2.96 ± 1.10	2.53 ± 0.76	2.34 ± 0.66	2.16 ± 0.60	0.16
∑ n-6 PUFA	6.64 ± 5.07	7.35 ± 4.33	8.56 ± 5.08	10.42 ± 6.37	0.41
n-3/n-6	0.57 ± 0.23 a	0.42 ± 0.15 ab	0.30 ± 0.07 b	0.25 ± 0.11 b	<0.01
C22:6n-3 (DHA)	1.42 ± 0.35 a	1.16 ± 0.33 ab	0.96 ± 0.12 bc	0.80 ± 0.17 c	<0.01
C20:5n-3 (EPA)	0.50 ± 0.25 a	0.43 ± 0.14 ab	0.35 ± 0.11 ab	0.27 ± 0.10 b	0.02
C20:4n-6 (ARA)	0.14 ± 0.02	0.13 ± 0.02	0.16 ± 0.05	0.14 ± 0.04	0.29
C22:5n-6 (DPA n-6)	0.10 ± 0.04	0.10 ± 0.05	0.12 ± 0.07	0.09 ± 0.03	0.60
DHA/EPA	3.09 ± 0.68	2.76 ± 0.50	2.89 ± 0.47	3.14 ± 0.84	0.52

SFAs = saturated fatty acids; MUFAs = monounsaturated fatty acids; PUFAs = polyunsaturated fatty acids; LC-PUFAs = long-chain polyunsaturated fatty acids; OA = oleic acid; LA = linoleic acid; GLA = Gamma-linolenic acid; ALA = Alpha-linolenic acid; DGLA = Dihomo-γ-linolenic acid; DHA = docosahexaenoic acid; EPA = eicosapentaenoic acid; ARA = arachidonic acid; and DPA = Docosapentaenoic acid. Significant differences are presented as different superscripts in each row (p < 0.05). Data are presented as mean ± SD of 10 replicates per dietary treatment. Statistical analysis was performed by one-way ANOVA, followed by Tukey’s multiple comparisons test.

). In addition, a higher inclusion of marine periphyton in the diet (16% FM and 0% FM diets) resulted in lower levels of EPA and DHA in the mullet fillet.

3.3. Diets’ Apparent Digestibility and Effects on the Fish Proximal Composition

The protein and lipid content measured in the fish tissue and feces samples were used to calculate each diet’s apparent digestibility of protein or lipids. The measured level of protein or lipids in the fish flesh under each examined diet in the digestibility trial was relatively similar to that measured in the feeding trial. Therefore, it can be assumed that the diet digestibility was similar in the different trials, ranging between an ADC of 71.04 and 82.01 for protein and 87.73 and 93.02 for lipids (

Table 7. Protein and lipid proximal composition in fillet and feces of Mugil cephalus, and the apparent digestibility coefficient (ADC) when they were fed the novel experimental diets.

Ingredients, % in DW	Diet
	Control	24% FM	16% FM	0% FM	p-Value
Fillet
Protein	63.7 ± 4.26	63.41 ± 3.02	61.64 ± 0.73	57.91 ± 3.13	0.16
Lipid	23.75 ± 4.86 a	22.27 ± 3.66 a	25.44 ± 1.18 a	37.90 ± 1.75 b	<0.01
Feces
Protein	22.59 ± 3.33	20.07 ± 1.88	22.96 ± 4.68	27.21 ± 11.10	0.60
Lipid	2.11 ± 0.04	1.81 ± 0.15	2.33 ± 0.43	2.56 ± 0.43	0.08
ADC
Protein	71.04 ± 9.27	78.30 ± 9.74	77.80 ± 11.74	82.01 ± 11.22	0.65
Lipid	87.73 ± 3.26	91.64 ± 3.76	90.24 ± 5.00	93.02 ± 2.43	0.41

All values are % of dry weight. Significant differences are presented as different superscripts in each row (p < 0.05). Data are presented as mean ± SD of 3 replicates per dietary treatment. Statistical analysis was performed by one-way ANOVA, followed by Tukey’s multiple comparisons test.

). The protein content in fish was relatively similar under the different diets (F3, 8 = 2.270, p = 0.16; Table 7). However, the lipid content in the fish fed the FM-free diet was 1.6 times higher than that in the control diet (F3, 8 = 14.83, p < 0.05; Table 7). In addition, the content of lipids or protein in the fish feces collected in the digestibility trial was relatively similar regardless of the provided diet (Table 7).

3.4. Diets’ Effect on Costs of Feed and Fish Production and FIFO Ratio

The control diet with 32% FM presented the highest cost, estimated at 0.73 USD kg⁻¹ (

Table 8. Economic analysis of the price of experimental diets and fish production, and the effects of the novel diets on the fish in:fish out ratio.

	Diet
	Control	24% FM	16% FM	0% FM
Feed cost, USD kg−1	0.73	0.58	0.49	0.42
Feed savings, %	-	19.75	32.77	42.63
Fish production cost, USD kg−1	2.72	2.27	1.897	2.094
FIFO ratio	5.3	4.1	2.8	-

Data for calculating feed prices were obtained from Zemach Feed Mill LTD or according to the price in the IndexMundi database on the day of aquafeed preparation. The cost of fish production was calculated for each experimental diet based on the cost of feed and the measured FCR (excluding dead fish) in the current study during 104 days of culture. Fish-in:fish-out (FIFO) ratio was calculated from current measurements of the input level of fish products (fishmeal and fish oil) in the formulated feed, amount of feed consumed (excluding dead fish), and fish yield.

). The cheapest diet was the FM-free feed, with an estimated cost of 0.42 USD kg⁻¹, equaling approximately 57% of the cost of the control aquafeed. The calculated savings in terms of feed prices in diets of 24% FM and 16% FM compared to the control diet were 19.75% and 32.77%, respectively. Although the FM-free aquafeed was the cheapest, the cost of fish production was lowest when feeding the fish the low-FM diet (16% FM), estimated at only 1.9 USD per each kg of produced fish biomass. The highest fish production cost of 2.72 USD kg⁻¹ was calculated for the control diet with 32% FM, followed by the low-periphyton diet (24% FM) with 2.27 USD kg⁻¹ (Table 8). As compared to the calculated FIFO of 5.3 for the control diet, the calculated FIFO ratios for the 24% FM and 16% FM diets were 4.1 or 2.8, respectively, or in other words, a savings of 23 or 48% in the required amount of fish for fish production with these diets (Table 8).

4. Discussion

The current study confirmed marine periphyton’s feasibility in the diet of grey mullet fingerlings. The inclusion of dietary periphyton is proposed for improving the sustainability of mullet production, mainly by reducing the FIFO ratio and feed and fish production costs. However, the supply of dietary EPA and DHA for grey mullets in high-salinity cultures is crucial due to the fish’s inability to synthesize these when their content in the aquafeed is insufficient, as occurs when eliminating both FM and FO.

The dietary requirements of grey mullets have not been fully addressed, but various experiments have reported their nutrition with FM substitutes. These have included soy [46,47], fermented plant feedstuffs [2], yeasts [48], and seaweeds [49,50]. Until now, periphyton in this fish diet was only as a supplemental feed accessible in culture ponds [27,29]. In the current study, periphyton was sufficient to halve the FM content in the aquafeed, but eliminating FM led to deficits in the fish growth and feed conversion to biomass. The measured FCR in the current study using periphyton agrees with those measured in a previous study of grey mullet fingerlings by Luzzana et al. (2005) [48], in which the experimental diets consisted of soybean meal or torula yeast to replace part of the fish and hemoglobin meal content to achieve an FCR of 4.3–4.9. Martínez-Antequera et al. (2022) [51] obtained relatively similar SGR and FCR values to those in the present study (10% FM and 3% FO) when feeding M. cephalus with a diet containing either a high level of fish products or one with only 13% fish products and a high (>75%) plant content. The high-plant diets resulted in an SGR between 0.2 and 0.5% d⁻¹ (compared to 0.75% in the control diet) and an FCR between 3.1 and 5.9 (compared to 2.19 in the control diet). This consistency is presumably due to the relatively similar conditions between their experiment and the current research, with their initial fish weight being approximately 12.2 vs. approximately 12.8 g in the current research and their culture salinity being approximately 37% vs. 40% in the current research. Another study revealed a relatively similar SGR of approximately 0.8–1% d⁻¹ by mullets when fed diets of low FM (6.5–13%) and FO (0.4–0.8%), but at a higher feeding rate than that in the current research [52]. An SGR of <1.1% d⁻¹ was also measured by Kalla et al. (2003) [46] when feeding mullets with plant-rich diets (>90% plants). In other studies that have examined diets with reduced FM content, grey mullets presented faster growth and a more efficient feed conversion [2,28] compared to the results of the FM-reduced diets in the current research. Those experiments revealed an SGR between 2.1% and 3% d⁻¹ and an FCR between 1.5 and 2.5 (2–4 times faster growth and a 1.5–2.5 times more efficient feed conversion to biomass). Such differences may be attributed to the different feeding regimes established in these studies, such as the higher feeding ratio of 10% of BW (2.5 times more feed than in the current study) [2] and the presence of additional natural feeds in the culture ponds [28,53]. In diets with a high-plant content, the fermentation of the plant ingredients before their integration into the aquafeed improves the digestibility of the feed’s macronutrients and, consequently, the fish growth. In the current study, the fish digestibility of the macronutrient protein and lipids was efficient even under the low-FM diets and even when compared to the control diet or previously reported values [46]. This suggests other factors may have caused the growth deficit in the here-examined mullets when fed the FM-free diet.

In this context, salinity has been reported to be a primary factor determining mullet growth [54,55]. High levels of salinity impaired grey mullets’ growth and digestion of feed while increasing the amount of feed required for them to gain weight [56,57]. Barman et al. (2005) [57] revealed a significant improvement in the fish SGR, feed conversion efficiency, and intestinal enzyme activity when reducing the culture salinity of grey mullets from 25‰ to 15‰ and then to 10‰. The salinity effect may be attributed to the energy costs of osmoregulation, which is minimal in mullets only under a particular level of salinity [54,58].

In contrast to the feed’s macronutrients, our analyses of the FA composition in fish tissue revealed deficits in the content of essential LC-PUFAs, particularly the n-3 PUFAs C22:6n-3 (DHA) and C20:5n-3 (EPA). The decreasing levels of both DHA and EPA in the fish tissue were associated with the decreasing FM content in the feed. Marine fish have a limited capacity for synthesizing EPA and DHA, which may lead to a reduced growth rate and feed efficiency [59,60]. Therefore, these essentials should be supplied through diet, as when integrating FO and FM into the formulated aquafeed. In the current study, the reduced FM content resulted in a reduction in EPA and DHA levels in the pelleted feed, while using plant oil (corn oil) did not compensate for such deficits.

Corn oil is considered a high-quality ingredient for fish diets, among other edible oils. It contains sufficient levels of saturated and unsaturated fatty acids of palmitoleic 16:1 (11.67%), stearic 18:0 (1.85%), oleic (25.16%), linoleic 18:2 (60.60%), linolenic 18:3 (0.48%), and ARA 20:0 (0.24%) [61]. However, as in other vegetable oils (e.g., flaxseed, canola), alpha-linolenic acid (ALA; C18:3n-3) is the primary omega-3 FA in corn oil, while DHA, EPA, and DPA are often absent.

Fish can convert ALA and other C18-unsaturated fatty acids, such as oleic acid (OA; C18:1n-9) and linoleic acid (LA; C18:2n-6), into EPA and DHA [62]. Although all three C:18 precursors were present in the currently examined feeds, we presume that the fish synthesis of EPA and DHA was negligible. This is demonstrated by the decreasing levels of EPA and DHA in the fish tissue when reducing the FM in their diet, while at the same time, the levels of the C:18 precursors in the tissues were similar regardless of diet type. Furthermore, the presence of C18:3n-6 (GLA) and C20:3n-6 (DGLA) in the tissue of mullets fed the different diets may suggest that these fish have the enzymes Δ8-desaturase and Δ8-elongase, which are involved in the biosynthesis of LC-PUFAs from the C18-precursors LA and ALA. The GLA and DGLA content was relatively similar in the diets but increased in tissues with decreased dietary FM levels. While eicosatrienoic acid (C20:4n-3) was absent from the experimental feeds examined here, it appeared in the mullets’ tissue. Eicosatrienoic acid is involved in the conversion of ALA (C18:3n-3) to EPA and, ultimately, DHA [63]. However, these end products of biosynthesis decreased in the fish tissue in proportion to their decrease in the examined feeds. Hence, our findings propose a partial metabolic pathway for LC-PUFAs towards the biosynthesis of EPA and DHA, which was not completed, presumably due to the poor activity of the required desaturases Δ5 and Δ4. The inability of marine fish to complete these metabolic pathways was associated with the presence of dietary omega-3 LC-PUFAs in their natural marine environment [64]. In partial agreement, Rosas et al. (2019) [65] reported on the elongation of C18:3n-3 to C20:3n-3 when providing mullets with diets with linseed oil and spirulina (Arthrospira platensis) as FO and FM substitutes, but a desaturation to C20:4 or C20:5n-3 was not obtained, resulting in a lower final weight of the fish when eliminating FM and FO in their diet.

Abiotic factors such as salinity and temperature are essential in determining fish’s requirements for FA [9]. For example, temperature influences the unsaturation of the phospholipid membrane, which is necessary for preserving membrane fluidity [66]. Therefore, omega-3 FAs are more necessary when the temperature declines, as they offer more unsaturation than omega-6 FA [67]. High salinity levels also elevate the need for omega-3 highly unsaturated FAs [9]. Altogether, these studies suggest that mullets’ requirement for specific FAs depend on their culture conditions due to variations in their LC-PUFA biosynthesis capability. While C18 precursors in the diet can be further metabolized in freshwater, high salinity levels will tip the balance toward the need to supply EPA and DHA through dietary raw materials. In the context of the diets examined here, the requirement for EPA and DHA was significant when reducing the FM content to 16% or lower. While corn oil efficiently supplied dietary FA for grey mullets in brackish water aquaculture [28], in the high-salinity culture, the lack of EPA and DHA in this ingredient was critical, as in the current study.

Marine periphyton in the current study was also short in terms of EPA and DHA. However, the enrichment of periphyton with diatoms may be efficient for increasing the LC-PUFA content in periphyton and consequently its nutritional value. This can be achieved by adding silica to the culture or the biofilter, as demonstrated previously [16,68].

Economically speaking, reducing the content of dietary FM from the aquafeed reduced the feed and fish production costs. Significant savings were documented when halving the level of FM in the aquafeed, resulting in feed and fish production costs that were only 70% of their costs in the control diet. Although greater savings were calculated for the FM-free diet, the profitability of such a diet for farmers is questionable due to the need to significantly extend the culture period until the desired fish weight for marketing is reached. With that being said, according to the Marine Ingredients Organization (IFFO), the FM yield from wild-caught fish is estimated at 22.5%, while the FO yield is estimated at 5%. Therefore, reducing the FM content in the formulated feeds reduces the amount of harvested wild fish, which should have a great ecologic and economic significance towards sustainable practices in mariculture. It should be noted, however, that the lack of FO in the current experimental diets had the greatest impact on the overall reduction in the use of wild-caught fish, while the partial replacement of FM with periphyton allowed for the further minimization of the exploitation of this natural resource. That being said, the FIFO ratios obtained for the 24% FM (4.1) and the 16% FM (2.8) diets in the current study are comparable to the FIFO ratios in diets with other substitutes for FM, such as soymeal (2.98), torula yeast (2.90), or Ulva (3.6) [28,48]. Among the reported diets, using oil seed meal protein resulted in the lowest FIFO ratio of 1.15 to 0.35 [53].

To conclude, the results of this study support the application of marine periphyton in aquafeed to supply dietary macronutrients for grey mullets. A partial, but still significant, reduction in the fishmeal content to 16% through the inclusion of periphyton improves mullet culture sustainability by the dual use of periphyton for biofiltration and nutrition, hence converting the excess nutrients in fishpond effluents into nutritionally valuable biomass. Moreover, the sustainability of the culture is also improved through the reduction in the FIFO ratio and savings in fish production costs. To further improve this technology, studies should aim for methods toimprove the content of protein and essential fatty acids in periphyton.