Lipid and Fatty Acid Composition of Low-Value Mediterranean Fish in Winter and Spring for Discard Valorization

Abstract

Winter and Spring variations of the fat and fatty acid compositions of discards from six species of the Aegean Sea were investigated to assess the potential suitability for human or aquaculture consumption. European pilchard (Sardina pilchardus), anchovy (Engraulis encrasicolus), curled picarel (Centracanthus cirrus), gilt sardine (Saridenella aurita), horse mackerel (Trachurus mediterraneus) and bogue (Boops boops) were collected from the local fishing wharf during winter and spring. In most species, the specimens caught in spring exhibited elevated fat and n-3 long-chain polyunsaturated fatty acid content, with Sardina pilchardus showing an increase in DHA from 13.59% to 16.06% and Engraulis encrasicolus from 20.36% to 23.41% of the total identified fatty acids. Despite their lower commercial value, the high n-3 LC-PUFA content renders them nutritionally valuable and eligible for use by the aquafeed industry as an alternative to the increasingly costly fish oil. Moreover, in accordance with EU legislation banning discards and mandating the landing of unwanted catches, the valorisation of these species, in line with circular economy principles, could enhance fishers’ income, reduce waste and contribute to the long-term sustainability of marine ecosystems.

1. Introduction

Marine capture fisheries contribute significantly to global food security by providing essential nutrients and economic development through employment, trade and the seafood supply chain. Yet a substantial portion of their yield is often discarded due to regulatory, economic, or operational constraints. Globally, discards represent not only a major loss of potentially valuable resources but also a challenge for sustainable fisheries management. According to FAO, discards account for 10.1% of the annual global catches (about 9.1 million tonnes) [1]. The Code of Conduct for Responsible Fisheries recommends that States should apply the precautionary approach widely to conservation management and exploitation of living aquatic resources, in order to protect them and preserve the aquatic environment [2]. Furthermore, States should take appropriate measures to minimize waste and discards, while maintaining relevant data [2]. In that context, the EU legislation (Regulation (EU) No 1380/2013) requires the landing and recording of all catches subject to catch limits, with the objective of gradually eliminating discards [3]. Moreover, landed discarded species are prohibited for human consumption and their utilization is restricted for other purposes such as fishmeal, fish oil, pet food, pharmaceuticals and cosmetics [3].

However, this policy—commonly referred to as the “Landing Obligation”—has sparked renewed interest in the valorization of underutilized or low-value fish resources. Following the global trend, approximately 240,000 tonnes of fish are discarded annually in the Mediterranean Sea, primarily consisting of species with low or no commercial value [4]. In the EU Regulation 1380/2013 definition, “discards” means catches that are returned to the sea [3]. There is no standard international definition of discards due to the national differences in how bycatch is characterized. However, discarded fish are typically identified based on species, size or a combination thereof, typically in relation to their economic value [5]. In the Mediterranean region, commercially important species such as Sardina pilchardus, Engraulis encrasicolus and Trachurus mediterraneus are often discarded when caught below the legal minimum landing size, while other species like Spicara smaris, Centracanthus cirrus and Boops boops are typically discarded due to their low or negligible market value [6,7].

Despite their low commercial value, discarded species retain high nutritional value [8], as they are sources of bioactive compounds such as protein, n-3 long chain polyunsaturated fatty acids (LC-PUFA), vitamins, polysaccharides and carotenoids [9,10,11]. These fatty acids are particularly important for human health, due to their anti-inflammatory, antithrombotic, antiarrhythmic, hypolipidemic and vasodilatory properties [12,13]. This underscores the nutritional relevance of these species, which have often been overlooked due to their limited market appeal. Their valorization could contribute not only to reducing waste but also to increasing the availability of healthy marine-derived lipids for food and feed purposes.

At the same time, the continuous growth of the aquaculture industry presents challenges regarding fish meal and fish oil availability [14,15]. Their increasing demand and the profitability of the sector have intensified the search for additional or alternative ingredients for aquafeed formulation [15]. A clearer understanding of the winter–spring fluctuations in these bioactive components is crucial for optimizing their nutritional utilization. In this context, small pelagic and demersal species discarded in the Mediterranean represent a promising yet underexplored source of n-3 LC-PUFA-rich biomass.

The fat content and fatty acid composition of harvested fish species, including those commonly discarded, exhibit marked winter–spring variations that can significantly influence their nutritional quality and potential applications [16,17,18,19]. These variations are related to the fish life cycle, temperature, salinity and the fatty acid composition of their natural diet [19]. Despite being recognized as excellent sources of n-3 LC-PUFAs, commonly discarded species in the Mediterranean Sea remain understudied. Even fewer studies have examined the seasonal variations in these compounds. Such information is necessary to identify optimal harvesting periods that maximize the yield of valuable nutrients and facilitate their valorization for feed or nutraceutical applications.

The aim of this study was to investigate the winter–spring variations of fat and fatty acid composition of discards from six species of the Aegean Sea—European pilchard (Sardina pilchardus), anchovy (Engraulis encrasicolus), curled picarel (Centracanthus cirrus), gilt sardine (Saridenella aurita), horse mackerel (Trachurus mediterraneus) and bogue (Boops boops)—and to assess their potential suitability as ingredients for processed products intended for human consumption or use in the aquaculture industry.

2. Materials and Methods

2.1. Sample Collection and Preparation

Specimens of the six study species—S. pilchardus, E. encrasicolus, C. cirrus, S. aurita, T. mediterraneus and B. boops—were collected from discards at the local fishing wharf in Keratsini (Piraeus) during winter (January) and spring (May) of 2015. Samples were obtained directly from landed discards, with no detailed catch location data available. Immediately after landing, individuals were sorted by species based on morphological characteristics following the FAO species identification guide for the Eastern Mediterranean [20]. From each crate, approximately 100 specimens of each species were measured and inspected. While individuals were broadly categorized as mature or immature during sorting, detailed records of maturity stages were not retained. Sex was not determined because whole-body samples were pooled and analyzed irrespective of sex. For each species and season, specimens were combined to create a composite sample of approximately 1 kg of homogenized whole-body tissue. Whole-body composites, rather than fillets, were analyzed to represent the total lipid content and fatty acid composition of discards, including muscle, skin, head and viscera. Samples were stored on ice and transported to the laboratory within 2 h. Fish were homogenized and subsamples were taken for moisture analysis in order to express crude lipid content on a wet weight basis. All analyses were performed in triplicate for each sample to ensure reproducibility and accuracy. Moisture was determined by placing the samples in an oven (110 °C for 24 h) until constant weight was obtained. The remaining homogenate was freeze-dried and stored at −20 °C until further analysis.

2.2. Crude Lipid Determination and Fatty Acid Composition Analysis

Crude lipid content was determined using the Folch extraction method [21] on homogenized whole-body composite samples. Fatty acid composition was analysed by gas chromatography (GC) after lipid extraction and subsequent methylation following Christie’s protocol [22]. Briefly, extracted lipids were saponified with 0.5 M KOH in methanol (Sigma-Aldrich, St. Louis, MO, USA), esterified with boron trifluoride–methanol (BF₃–MeOH; Sigma-Aldrich, St. Louis, MO, USA) at 100 °C for 30 min. The resulting FAMEs were extracted in hexane (Sigma-Aldrich, St. Louis, MO, USA), washed with distilled water and dried under nitrogen. Fatty acid methyl esters (FAMEs) were separated using a fused silica capillary column, coated with bonded polyglycol liquid phase (Omegawax 320, L: 30 m, ID: 0.32 mm, DF: 0.25 μm; Supelco, Bellefonte, PA, USA), on an Agilent GC (6890) system, coupled with a flame ionization detector (FID) (Agilent Technologies, Santa Clara, CA, USA).

Helium (99.999% purity; Revival, Athens, Greece) was used as the carrier gas, with a flow rate of 2 mL/min. The temperature program was set from 50 °C to 150 °C at a rate of 40 °C/min and then increased to 225 °C at a rate of 2 °C/min. Nonadecanoic acid (19:0; Sigma-Aldrich, St. Louis, MO, USA) was used as the internal standard. FAME identification was performed by comparison to known standards and reference materials (37 Components FAME mix, Menhaden Fish Oil and AOCS Ce 1i-07; Supelco, Bellefonte, PA, USA) using Chemstation software (G1701DA Rev.01.02 SP1; Agilent Technologies, Santa Clara, CA, USA).

2.3. Statistical Analysis and Modeling

Statistical analysis was performed using R (version 4.5.1). Differences in fatty acid composition between winter and spring samples were evaluated for each species using independent samples t-tests. Prior to testing, data were checked for normality using the Shapiro–Wilk test and homogeneity of variances using Levene’s test. A significance threshold of p < 0.05 was applied.

To explore multivariate patterns of variation, Principal Component Analysis (PCA) was conducted on the major dietary fatty acids. The data were autoscaled (mean-centered and standardized) prior to analysis. PCA scores were plotted to visualize sample groupings by species and season, while loading vectors were used to identify the fatty acids contributing most to the observed variation. The analysis enabled the identification of winter–spring shifts in lipid profiles and provided insight into the dominant patterns across species. The PCA results were interpreted in conjunction with univariate outcomes to draw robust conclusions about the winter–spring variability and its biological significance.

In addition to univariate analyses, a two-way multivariate analysis of variance (MANOVA) was performed to examine the simultaneous effects of species, season and their interaction on the fatty acid profile. MANOVA was chosen to account for intercorrelations among fatty acids and detect multivariate patterns. Box’s M test indicated heterogeneity of covariance matrices (p < 0.001), so Pillai’s Trace was used as the criterion for significance. Follow-up univariate ANOVAs were conducted to identify specific fatty acids contributing to multivariate effects. Approximately 100 specimens per species and season (winter and spring) were initially measured and sorted, but analyses were performed on pooled composites (one per species and season), each analyzed in triplicate. Therefore, the results reflect pooled inter-individual variability and are interpreted accordingly.

3. Results

3.1. Crude Lipid Content

Crude lipid content varied significantly between winter and spring for most of the examined species (

Table 1. Crude lipid content (% wet weight ± standard deviation, n = 3 analytical replicates) of specimens collected during winter and spring.

	Winter	Spring	p
Sardina pilchardus	4.0 ± 0.08	11.0 ± 0.23	***
Engraulis encrasicolus	2.0 ± 0.11	2.5 ± 0.13	*
Centracanthus cirrus	1.7 ± 0.09	7.2 ± 0.10	*
Saridenella aurita	5.5 ± 0.29	5.2 ± 0.11	NS
Trachurus mediterraneus	2.9 ± 0.13	5.6 ± 0.33	***
Boops boops	1.9 ± 0.07	8.4 ± 0.23	***

(NS: not significant; *: p < 0.05; ***: p < 0.001).

). In general, specimens collected in spring exhibited higher fat content compared to those caught in winter. The most pronounced winter–spring differences were observed in Sardina pilchardus, Centracanthus cirrus, Trachurus mediterraneus and Boops boops (p < 0.001), with lipid levels increasing from 4.0% to 11.0%, 1.7% to 7.2%, 2.9% to 5.6% and 1.9% to 8.4%, respectively. A smaller but statistically significant increase was observed in Engraulis encrasicolus (p < 0.05). In contrast, Saridenella aurita showed no significant winter–spring variation in lipid content (p > 0.05).

3.2. Fatty Acid Composition

The fatty acid composition also showed winter–spring variation, with significant differences between winter and spring in several key fatty acids across species (

Table 2. Selected dietary fatty acid profile (% total identified fatty acids ± standard deviation, n = 3 analytical replicates) of European pilchard (Sardina pilchardus), anchovy (Engraulis encrasicolus), curled picarel (Centracanthus cirrus) specimens collected during winter and spring.

	Sardina pilchardus			Engraulis encrasicolus			Centracanthus cirrus
	Winter	Spring	p	Winter	Spring	p	Winter	Spring	p
14:0	5.69 ± 0.316	5.99 ± 0.055	NS	2.77 ± 0.129	3.32 ± 0.122	**	3.23 ± 0.233	3.10 ± 0.151	NS
16:0	18.49 ± 0.742	16.99 ± 0.251	*	22.58 ± 0.560	21.77 ± 1.059	NS	19.30 ± 0.518	21.50 ± 0.648	*
16:1n-7	7.32 ± 0.249	6.54 ± 0.131	**	3.51 ± 0.120	3.33 ± 0.116	NS	3.81 ± 0.145	4.71 ± 0.123	**
17:0	0.83 ± 0.014	0.54 ± 0.014	***	0.00 ± 0.000	0.45 ± 0.022	***	0.31 ± 0.005	0.16 ± 0.016	***
18:0	4.00 ± 0.322	3.58 ± 0.097	NS	5.71 ± 0.316	5.17 ± 0.150	NS	6.88 ± 0.229	5.82 ± 0.332	*
18:1n-9	5.20 ± 0.248	8.06 ± 0.065	***	7.65 ± 0.065	7.45 ± 0.406	NS	8.45 ± 0.161	16.38 ± 0.521	***
18:1n-7	2.78 ± 0.130	2.12 ± 0.008	**	2.76 ± 0.070	2.41 ± 0.126	*	2.10 ± 0.021	2.49 ± 0.071	**
18:2n-6	0.98 ± 0.021	1.25 ± 0.023	***	1.37 ± 0.067	1.88 ± 0.094	**	1.73 ± 0.019	1.06 ± 0.027	***
18:3n-3	0.64 ± 0.000	1.10 ± 0.024	***	0.97 ± 0.015	1.77 ± 0.068	***	1.11 ± 0.049	0.74 ± 0.015	***
18:4n-3	2.08 ± 0.051	3.28 ± 0.082	***	2.16 ± 0.029	2.49 ± 0.137	*	1.63 ± 0.200	0.77 ± 0.021	**
20:0	0.32 ± 0.008	0.48 ± 0.013	***	0.58 ± 0.085	0.66 ± 0.053	NS	0.99 ± 0.070	0.23 ± 0.019	***
20:1n-11 + 20:1n-9	5.62 ± 0.448	5.31 ± 0.271	NS	0.69 ± 0.050	1.13 ± 0.310	NS	1.20 ± 0.172	1.91 ± 0.091	**
20:2n-6	0.08 ± 0.032	0.00 ± 0.000	*	ND	ND	NS	0.00 ± 0.000	0.06 ± 0.002	***
20:4n-6	0.21 ± 0.108	0.24 ± 0.022	NS	ND	ND	NS	0.00 ± 0.000	0.04 ± 0.003	***
20:4n-3	1.02 ± 0.362	0.47 ± 0.023	NS	1.30 ± 0.081	0.79 ± 0.062	**	1.50 ± 0.042	0.93 ± 0.105	**
20:5n-3	7.95 ± 0.213	7.17 ± 0.094	**	8.03 ± 0.294	7.09 ± 0.306	*	6.54 ± 0.051	5.39 ± 0.288	**
22:1n-11	7.23 ± 0.866	5.95 ± 0.401	NS	0.35 ± 0.070	0.85 ± 0.032	***	0.18 ± 0.024	0.24 ± 0.012	*
22:1n-9	0.38 ± 0.025	0.40 ± 0.003	NS	0.21 ± 0.058	0.22 ± 0.045	NS	0.25 ± 0.017	0.24 ± 0.036	NS
22:5n-3	1.31 ± 0.160	1.62 ± 0.244	NS	1.22 ± 0.144	1.24 ± 0.411	NS	1.18 ± 0.042	2.54 ± 0.301	**
22:6n-3	13.59 ± 0.954	16.06 ± 0.193	*	20.36 ± 0.990	23.41 ± 1.154	*	22.03 ± 0.218	18.37 ± 1.105	**
24:1n-9	1.47 ± 0.144	1.45 ± 0.126	NS	1.71 ± 0.025	2.35 ± 0.563	NS	2.33 ± 0.040	1.84 ± 0.329	NS
SFA	29.33 ± 1.294	27.57 ± 0.304	NS	31.64 ± 0.929	31.36 ± 1.321	NS	30.71 ± 0.883	30.80 ± 1.028	NS
MUFA	30.01 ± 1.606	29.82 ± 0.669	NS	16.88 ± 0.046	17.73 ± 0.583	NS	18.31 ± 0.231	27.80 ± 0.929	***
Σn-3	26.59 ± 1.589	29.68 ± 0.188	*	34.04 ± 1.391	36.80 ± 1.332	NS	33.98 ± 0.458	28.74 ± 1.384	**
Σn-3 LC PUFA	23.87 ± 1.558	25.31 ± 0.293	NS	30.91 ± 1.386	32.53 ± 1.173	NS	31.24 ± 0.230	27.23 ± 1.349	**
Σn-6	1.27 ± 0.152	1.49 ± 0.017	NS	1.37 ± 0.067	1.88 ± 0.094	**	1.73 ± 0.019	1.16 ± 0.031	***
EPA + DHA	21.54 ± 1.166	23.23 ± 0.197	NS	28.39 ± 1.281	30.51 ± 1.452	NS	28.56 ± 0.226	23.75 ± 1.383	**
SFA/MUFA	0.98 ± 0.038	0.92 ± 0.030	NS	1.87 ± 0.052	1.77 ± 0.054	NS	1.68 ± 0.033	1.18 ± 0.164	**
Σn-3/Σn-6	20.95 + 1.215	19.92 + 0.355	NS	24.84 + 0.205	19.54 + 0.710	***	19.70 + 0.459	24.87 + 0.902	**

LC PUFA: long-chain polyunsaturated fatty acids; SFA: saturated fatty acids; MUFA: monounsaturated fatty acids. (ND: Not Detected (below instrument detection limit); NS: not significant; *: p < 0.05; **: p < 0.01; ***: p < 0.001).

). In Sardina pilchardus, spring samples had higher levels of 18:1n-9 (oleic acid), 18:2n-6 (linoleic acid), 18:3n-3 (α-linolenic acid), 18:4n-3 (stearidonic acid) and 22:6n-3 (DHA), with concurrent decreases in 16:0 (palmitic acid), 16:1n-7 (palmitoleic acid) and 18:1n-7 (vaccenic acid). Specifically, DHA increased from 13.59% in winter to 16.06% in spring. EPA (20:5n-3) was also found to be significantly lower in the spring. Despite this reduction, the total n-3 fatty acid content was significantly elevated in spring, compared to winter.

Similar patterns were observed in Engraulis encrasicolus, with DHA increasing from 20.36% to 23.41%; however, the increase in total n-3 fatty acid level did not reach statistical significance. This is because Σn-3 is the sum of multiple n-3 components with varying winter–spring trends and the combined variability offsets the significant increase observed in individual fatty acids such as DHA. In contrast, spring levels of palmitic, oleic and vaccenic acid in Centracanthus cirrus were elevated, while EPA, DHA, linoleic, α-linolenic and stearidonic acid levels declined. Additionally, total n-3, n-3 LC-PUFA and n-6 levels decreased, whereas total monounsaturated fatty acids increased.

In all three species, DHA and EPA remained the dominant n-3 polyunsaturated fatty acids and palmitic acid was the most abundant saturated fatty acid. Notably, the Σn-3/Σn-6 ratio significantly decreased in E. encrasicolus (p < 0.001) and increased in C. cirrus (p < 0.01), indicating species-specific shifts in nutritional profile.

For the remaining species, winter–spring trends were more variable (

Table 3. Selected dietary fatty acid profile (% total identified fatty acids ± standard deviation, n = 3 analytical replicates) of gilt sardine (Saridenella aurita), horse mackerel (Trachurus mediterraneus) and bogue (Boops boops) specimens collected during winter and spring.

	Saridenella aurita			Trachurus mediterraneus			Boops boops
	Winter	Spring	p	Winter	Spring	p	Winter	Spring	p
14:0	6.84 ± 0.117	4.96 ± 0.238	***	3.70 ± 0.072	3.04 ± 0.061	***	5.69 ± 0.086	5.54 ± 0.156	NS
16:0	21.17 ± 0.911	19.98 ± 0.758	NS	19.56 ± 0.397	20.44 ± 0.227	*	20.23 ± 0.169	21.92 ± 0.262	**
16:1n-7	6.02 ± 0.035	5.02 ± 0.223	**	5.65 ± 0.182	4.92 ± 0.148	**	7.32 ± 0.155	6.97 ± 0.080	*
17:0	0.54 ± 0.014	0.36 ± 0.009	***	0.23 ± 0.050	0.13 ± 0.031	NS	0.26 ± 0.025	0.19 ± 0.014	*
18:0	5.51 ± 0.275	5.13 ± 0.178	NS	6.05 ± 0.178	6.68 ± 0.280	*	7.10 ± 0.182	6.96 ± 0.162	NS
18:1n-9	6.53 ± 0.244	6.99 ± 0.437	NS	19.72 ± 0.255	19.70 ± 0.858	NS	16.33 ± 0.125	16.56 ± 0.042	*
18:1n-7	2.60 ± 0.062	2.43 ± 0.081	*	2.77 ± 0.105	2.92 ± 0.020	NS	3.28 ± 0.087	3.20 ± 0.018	NS
18:2n-6	1.86 ± 0.119	1.58 ± 0.046	*	1.19 ± 0.017	1.24 ± 0.023	*	1.24 ± 0.046	1.19 ± 0.033	NS
18:3n-3	1.14 ± 0.015	1.05 ± 0.052	*	0.87 ± 0.053	0.80 ± 0.022	NS	0.89 ± 0.006	0.80 ± 0.008	***
18:4n-3	1.82 ± 0.038	1.64 ± 0.070	*	0.98 ± 0.026	0.96 ± 0.056	NS	1.10 ± 0.101	1.08 ± 0.034	NS
20:0	0.71 ± 0.135	0.34 ± 0.013	**	0.34 ± 0.012	0.35 ± 0.025	NS	0.54 ± 0.025	0.48 ± 0.050	NS
20:1n-11 + 20:1n-9	0.70 ± 0.040	0.64 ± 0.085	NS	2.28 ± 0.060	2.00 ± 0.103	*	1.82 ± 0.051	1.65 ± 0.079	*
20:2n-6	0.07 ± 0.029	ND	*	0.00 ± 0.000	0.02 ± 0.029	NS	0.00 ± 0.000	0.02 ± 0.040	NS
20:4n-6	0.08 ± 0.049	ND	*	0.00 ± 0.000	0.06 ± 0.048	NS	0.00 ± 0.000	0.01 ± 0.024	NS
20:4n-3	1.74 ± 0.059	1.39 ± 0.021	**	1.70 ± 0.046	1.34 ± 0.074	**	1.61 ± 0.040	1.41 ± 0.055	**
20:5n-3	9.82 ± 0.461	9.81 ± 0.364	NS	5.57 ± 0.200	5.20 ± 0.177	NS	5.18 ± 0.045	4.86 ± 0.254	NS
22:1n-11	0.09 ± 0.024	0.36 ± 0.149	*	0.58 ± 0.093	0.81 ± 0.029	*	0.19 ± 0.015	0.22 ± 0.021	NS
22:1n-9	0.24 ± 0.012	0.20 ± 0.023	NS	0.33 ± 0.015	0.34 ± 0.010	NS	0.43 ± 0.047	0.45 ± 0.025	NS
22:5n-3	1.56 ± 0.775	2.23 ± 0.759	NS	2.04 ± 0.071	2.15 ± 0.152	NS	1.93 ± 0.055	2.16 ± 0.325	NS
22:6n-3	16.48 ± 1.324	22.10 ± 0.488	**	13.71 ± 0.392	15.65 ± 0.461	**	10.04 ± 0.067	10.72 ± 0.549	NS
24:1n-9	1.21 ± 0.504	1.57 ± 0.543	NS	1.01 ± 0.035	1.22 ± 0.068	**	0.80 ± 0.035	0.89 ± 0.029	*
SFA	34.77 ± 1.096	30.77 ± 1.108	*	29.87 ± 0.499	30.66 ± 0.170	NS	33.81 ± 0.265	35.08 ± 0.533	*
MUFA	17.39 ± 0.495	17.21 ± 0.525	NS	32.33 ± 0.319	31.92 ± 1.050	NS	30.18 ± 0.196	29.94 ± 0.245	NS
Σn-3	32.55 ± 1.386	38.22 ± 0.895	**	24.87 ± 0.121	26.10 ± 0.870	NS	20.75 ± 0.282	21.04 ± 0.575	NS
Σn-3 LC PUFA	29.60 ± 1.342	35.53 ± 1.014	**	23.02 ± 0.196	24.34 ± 0.800	NS	18.76 ± 0.187	19.15 ± 0.589	NS
Σn-6	2.01 ± 0.041	1.58 ± 0.046	***	1.19 ± 0.017	1.32 ± 0.046	*	1.24 ± 0.046	1.23 ± 0.082	NS
EPA + DHA	26.30 ± 1.766	31.91 ± 0.379	**	19.28 ± 0.238	20.85 ± 0.635	*	15.21 ± 0.093	15.58 ± 0.801	NS
SFA/MUFA	2.00 ± 0.088	1.79 ± 0.069	*	0.92 ± 0.014	0.96 ± 0.027	NS	1.12 ± 0.007	1.17 ± 0.017	**
Σn-3/Σn-6	16.16 + 0.679	24.14 + 1.221	**	20.91 + 0.388	19.85 + 1.134	NS	16.74 + 0.402	17.16 + 1.419	NS

LC PUFA: long-chain polyunsaturated fatty acids; SFA: saturated fatty acids; MUFA: monounsaturated fatty acids. (ND: Not Detected (below instrument detection limit); NS: not significant; *: p < 0.05; **: p < 0.01; ***: p < 0.001).

). Sardinella aurita demonstrated significant reductions in most fatty acid spring levels. Notably, DHA (p < 0.01) and 22:1n-11 (p < 0.05) were exceptions, showing significantly higher concentrations in spring. The increase in DHA also contributed to significant elevations in total n-3 and total n-3 LC-PUFA.

Horse mackerel also showed significant increases in DHA and total Σn-3 content in spring, while Boops boops presented relatively stable profiles, with only minor differences between winter and spring.

Overall, the winter–spring variation in fatty acid profiles was more pronounced in species with higher seasonal lipid accumulation, suggesting a relationship between total fat content and compositional shifts. Long-chain n-3 LC-PUFA levels (particularly EPA + DHA) remained high across all species and seasons, confirming the potential nutritional value of these low-commercial species.

3.3. Principal Component Analysis

The results of the principal component analysis (PCA) were in strong agreement with the univariate statistical analysis of fatty acid composition. Species such as Sardina pilchardus and Trachurus mediterraneus showed clear separation between winter and spring along PC1, which was driven by fatty acids that also exhibited statistically significant winter–spring differences (e.g., DHA, 18:1n-9, and α-linolenic acid). Engraulis encrasicolus displayed moderate separation, reflecting directionally similar but less statistically robust changes. In contrast, Boops boops and Sardinella aurita formed tight seasonal clusters in the PCA plot and showed few or no significant differences in individual fatty acids, indicating minimal winter–spring variability. Notably, Centracanthus cirrus exhibited a distinct winter–spring shift in the opposite direction along PC1, associated with increases in SFA and MUFA during spring, aligning with significant univariate changes. A summary of the agreement between PCA-based grouping and the results of t-tests for each species is presented in

Table 4. Summary of PCA and t-test agreement by species.

Species	PCA Separation (Winter vs. Spring)	t-Test Support	Interpretation
Sardina pilchardus	Clear (PC1)	Strong	Pronounced winter–spring variation
Engraulis encrasicolus	Moderate	Partial	Directional winter–spring trend, less robust
Centracanthus cirrus	Clear, opposite shift	Strong	Unique winter–spring dynamics
Sardinella aurita	None	Weak/None	Stable fatty acid profile
Trachurus mediterraneus	Clear (PC1)	Strong	Strong winter–spring variation
Boops boops	Overlapping	Weak/None	Minimal winter–spring difference

.

The first two principal components explained 56.9% of the total variance after Varimax rotation, while six components with eigenvalues greater than 1 accounted for 90.5% of the variation in fatty acid profiles. The rotated solution redistributed variance more evenly across components, allowing for improved interpretation of species- and season-specific patterns (Figure 1).

3.4. Multivariate Analysis of Variance

The MANOVA results revealed a significant multivariate effect of species (Pillai’s Trace = 4.980, F (115, 30) = 66.63, p < 0.001), season (Pillai’s Trace = 0.999, F (23, 2) = 58.29, p = 0.017) and their interaction (Pillai’s Trace = 4.917, F (115, 30) = 15.45, p < 0.001) on overall fatty acid profiles. Follow-up univariate ANOVAs showed that several individual fatty acids were significantly affected by species, season, or their interaction (Figure 2). Key winter–spring changes included elevated spring levels of DHA, EPA and Σn-3 in multiple species. Although the primary statistical focus was on winter–spring changes, the MANOVA results also highlight significant differences across species, while the PCA biplot provides a clear visual representation of inter-species similarities and groupings based on fatty acid composition. The significant species–season interaction detected in the MANOVA is reflected in the species-specific lipid shifts shown in Table 2 and Table 3, for instance, DHA increases in spring for E. encrasicolus and T. mediterraneus, while an opposite trend is observed in C. cirrus. This highlights how multivariate effects translate into individual fatty acid patterns.

4. Discussion

The findings of this study contribute to the utilization potential of several low-commercial value fish species, commonly discarded in the Aegean Sea, as they exhibit notable winter–spring variation in their lipid content and fatty acid profiles. Crude lipid levels were significantly higher in spring for all species examined, with the exception of the gilt sardine. This pattern is consistent with previous studies linking seasonal fat accumulation to post-spawning recovery and increased food availability during the warmer months [17,18,19,23,24]. The significant decrease in EPA observed in sardine and anchovy during spring, alongside an increase in DHA, may reflect the selective oxidation of EPA and the preferential retention of DHA, likely due to the inherent difficulty in oxidizing 22:6n-3 [25,26]. Other LC-PUFA, such as docosapentaenoic acid (22:5n-3), also displayed winter–spring variation, particularly in species like C. cirrus, where spring levels increased significantly (Table 2). Although less abundant than EPA and DHA, 22:5n-3 contributes to the overall nutritional profile of these species [27]. EPA (20:5n-3) is more prone to β-oxidation and is preferentially mobilized as an energy source during post-spawning recovery, whereas DHA (22:6n-3) is structurally essential in cell membranes and is thus selectively retained to preserve membrane integrity and function [25,26].

Moreover, the winter–spring variability of the fatty acid profile has been previously attributed to the corresponding changes in the natural diet, such as fluctuations in the availability and composition of zooplankton, phytoplankton and other prey items that influence the lipid intake of fish [28,29,30]. An overview of the main prey items for each species during winter and spring, compiled from published studies, is provided in

Table A1. Main prey items of the studied species during winter and spring (based on the literature data).

Species (Scientific Name)	Winter Prey Items	Spring Prey Items	References
European pilchard (Sardina pilchardus)	Larvae (<10 mm SL): tintinnid protozoans (e.g., Codonellopsis sp., ~48%), copepod nauplii (~46%). Larger larvae (10–16 mm SL): calanoid copepod post-nauplii (mainly Clausocalanus spp., ~53%), copepod nauplii (~22%), plus cyclopoid and harpacticoid copepods, particulate organic matter, protozoan cysts	Copepods (e.g., Clausocalanus, Oncaea, Paracalanus), crustacean larvae, teleost eggs, dinoflagellates (Ceratium, Lingulodinium), tintinnids	[54,55,56]
Anchovy (Engraulis encrasicolus)	Diets of larvae (<9 mm SL) are dominated by copepod nauplii and tintinnid protozoans, with contributions from appendicularians and small cladocerans. Adults and juveniles feed mainly on copepods (Oncaea mediterranea, O. venusta, Microsetella rosea) and decapod larvae.	Larger larvae (>9 mm SL) and juveniles shift towards copepod post-nauplii (e.g., Clausocalanus, Candacia, Temora spp.), cladocerans (Evadne spp.), euphausiids and fish eggs. Seasonal shifts correspond to changes in zooplankton availability and spawning-related energy demands.	[56,57,58]
Curled picarel (Centracanthus cirrus)	Primarily small zooplankton—mainly copepods—along with occasional mysid shrimps and fish larvae.	Diet remains consistent year-round, dominated by copepods, mysids and fish larvae *	[59]
Gilt sardine (Sardinella aurita)	Primarily zooplanktonic crustaceans (mainly copepods), along with euphausiids, amphipods, decapod larvae, teleost eggs and larvae and siphonophores. Small individuals feed mostly on copepods and other microplankton.	Diet shifts to include larger zooplankton such as hyperiid amphipods, mysids, euphausiids, siphonophores and teleost larvae and eggs, while copepods remain important prey. Larvae (<8 mm) feed mainly on copepod nauplii and Evadne spp., whereas larger larvae consume more copepod postnauplii.	[34,35]
Horse mackerel (Trachurus mediterraneus)	Diet mainly consists of copepods (e.g., Corycaeus sp., Oncaea media, Euterpina acutifrons, Oithona nana), decapod larvae, bivalve larvae and small teleost larvae.	Increased consumption of polychaetes (Neanthes fucata, Platynereis dumerilii), decapod larvae, amphipods, isopods and euphausiids, with copepods still dominant. Fish eggs and larvae also contribute during spawning periods.	[60,61,62]
Bogue (Boops boops)	Omnivorous diet dominated by crustaceans (e.g., copepods), benthic organisms, small mollusks and seagrass fragments.	Similar diet with increased occurrence of gelatinous zooplankton (e.g., Pelagia noctiluca), fish larvae and seasonal phytoplankton.	[36,37,63]

* Note: The diet of Centracanthus cirrus is less extensively documented compared to other species in this study. The prey composition listed here is based on limited gut-content studies and general observations of small pelagic fish feeding habits.

(Appendix A). In sardine, total n-3 content was elevated in spring, consistent with previous findings [17,23,29]. While Sardinella aurita did not show a significant change in total lipid content between winter and spring, a notable springtime increase in total n-3 LC-PUFA was observed, in line with past reports [28,30]. Similar but not statistically significant trends were observed in the other species, except for curled picarel, which showed a slight decrease. In C. cirrus, the spring decrease in n-3 LC-PUFA may reflect post-spawning lipid mobilization, during which energy reserves are depleted. This pattern could also be linked to dietary shifts toward prey items with a lower n-3 content, as reported for other Mediterranean species [16,19,24]. EPA and DHA levels in anchovies were significantly higher in spring, consistent with previous data [19,29]. These elevated n-3 LC-PUFA concentrations across multiple underutilized species reinforce their value as nutrient-rich resources, despite their traditional classification as discards.

Further highlighting species-specific patterns, bogue has been previously reported to exhibit significant seasonal variations of EPA, DHA, n-3 LC-PUFA, SFA and MUFA [31,32,33]. However, such variations were not evident in the current study, likely due to differences in sampling periods, local environmental factors, or the specific fatty acid composition of the local zooplankton community [29]. In contrast, the relatively stable fatty acid profile of horse mackerel across seasons is consistent with earlier findings, which reported minimal fluctuations in its lipid composition between winter and spring [32]. This stability may also be related to the trophic flexibility of Sardinella aurita, which readily alternates between zooplankton and phytoplankton resources depending on availability [34,35]. In contrast, Boops boops shows an omnivorous feeding strategy, exploiting both pelagic and benthic prey, such as crustaceans, mollusks and gelatinous zooplankton, throughout the year [36,37]. These ecological traits likely reduce the impact of seasonal changes on lipid profiles. Comparable patterns of winter–spring lipid variation have been observed in other Mediterranean small pelagic and demersal species [16,19,23,24].

Taken together, these findings demonstrate that while some discarded species exhibit strong winter–spring shifts in lipid composition, others maintain a relatively stable profile throughout the year. Across all species, n-3 LC-PUFA content and Σn-3/Σn-6 ratios were high, with values higher or comparable to those found in commercial fish oils for use in aquaculture [38] and for human consumption [39]. The modern Western diet is characterized by a disproportionately high n-6/n-3 ratio, estimated at 10:1 to 20:1 [40,41], whereas a ratio closer to 5:1 is recommended for optimal health [42]. In this context, the fatty acid profiles of the studied species make them particularly suitable for inclusion in functional foods or aquafeeds. The presence of substantial levels of EPA and DHA further enhances their functional value.

To better capture the complexity of winter–spring lipid variability, the study employed a combination of univariate and multivariate statistical methods. The integration of multivariate (PCA and MANOVA) and univariate (t-test) analyses provides a more comprehensive understanding of winter–spring changes in fatty acid composition across species. Principal Component Analysis (PCA) effectively visualized patterns of variance in the data and allowed for the identification of species- and season-specific lipid signatures [28,29,43]. Species such as Sardina pilchardus and Trachurus mediterraneus showed clear winter–spring separation along PC1, which was driven by fatty acids that also exhibited significant winter–spring differences in univariate tests (e.g., DHA, 18:1n-9, α-linolenic acid) [17]. Engraulis encrasicolus showed moderate separation, suggesting less pronounced but directionally similar lipid shifts. Conversely, Boops boops and Sardinella aurita displayed tight winter–spring clustering, indicating low lipid variability, consistent with their ecological traits and diet. Interestingly, Centracanthus cirrus showed a shift in the opposite direction along PC1, driven by increases in saturated and monounsaturated fatty acids—highlighting species-specific physiological or trophic responses.

In parallel, Multivariate Analysis of Variance (MANOVA) further validated these patterns, detecting significant effects of species, season and their interaction on the overall fatty acid matrix. This analytical approach confirmed that the winter–spring trends observed in PCA were not merely visual artifacts but statistically robust across multiple variables [29,44]. The agreement between PCA visualization and MANOVA outcomes demonstrates the added interpretive value of combining dimension-reduction and hypothesis-testing frameworks when assessing biochemical variation in marine organisms [45,46,47]. Such an approach is particularly relevant in the context of discard utilization, where both ecological and biochemical profiles inform the potential for valorization [25,48].

One of the most immediate applications for valorizing these species is in aquafeed production. As global aquaculture continues to grow, the demand for fishmeal and fish oil—especially those rich in EPA and DHA—continues to rise, creating supply shortages and price volatility [14,15]. The inclusion of discarded species with favorable fatty acid composition in feeds can reduce dependency on small pelagic species, which are under constant pressure from overfishing due to their extensive use as raw materials for fishmeal and fish oil [49]. The inclusion of such species in feed formulations has already been shown to support optimal growth performance and health in farmed fish, making their integration both nutritionally feasible and economically relevant [8]. Although direct yield and cost comparisons with traditional sources such as anchovy or menhaden are limited, fishmeal and fish oil derived from discards have been shown to match the nutritional quality of industrial fishmeal and successfully support aquafeed formulations for species such as salmon, trout and seabream [8]. Since these species are typically treated as low-value byproducts, their acquisition cost is significantly lower than that of targeted forage fish, making them a promising alternative for sustainable feed production. This highlights a clear valorization pathway for these underutilized species as cost-effective raw materials for aquafeed and nutraceutical industries [8,9,10].

Beyond aquafeeds, human nutrition represents another important valorization route. Several of the studied species displayed Σn-3/Σn-6 ratios within or exceeding recommended dietary thresholds, indicating potential as sources of health-promoting lipids. Products derived from discarded fish—such as oil capsules, fortified foods, or protein–lipid concentrates—can serve as functional ingredients, addressing current gaps in dietary EPA and DHA intake, particularly in Western populations [39,42]. Although current regulations restrict the direct marketing of landed discards for human consumption [3] processed forms, such as purified oils and nutraceuticals, derived from these species are both legally permissible and technically feasible, especially when they are not marketed as whole foods [9,11,50,51].

Additionally, the pharmaceutical, cosmetic and pet food industries may represent valuable end-markets. Fish oils rich in LC-PUFA are increasingly used in formulations targeting inflammatory and cardiovascular conditions, while protein hydrolysates from marine discards are being investigated for anti-hypertensive and antioxidant properties [8,11]. The high-quality lipid profiles of the studied species align well with these trends, especially considering the relatively low contaminant burden often reported for small pelagics and short-lived demersal species [52].

Importantly, the winter–spring variation documented in this study underscores the need for strategic timing in harvesting and processing, in order to maximize lipid yield and nutritional value. This is particularly relevant when valorization is targeted toward high-end markets, such as supplements and pharmaceuticals, where compositional standards are more stringent. Year-round biochemical profiling of these species could help define optimal landing windows and enhance the economic efficiency of processing operations.

Reintegrating these low-value species into the value chain, either as feed ingredients or for the extraction of bioactive lipids, aligns with the objectives of the EU’s Landing Obligation [3]. The reduced discards and bycatch, enforced by changes in legislation or higher market prices, are expected to be a driving factor in the anticipated increase in capture fisheries production by 2032 [15]. This study supports the utilization of discards by demonstrating their nutritional potential, which is expected to contribute both to the income diversification of fishers and to the broader goal of sustainable fisheries management [53].

The data were obtained from samples collected and analyzed in 2015, while the subsequent application of multivariate methods (PCA and MANOVA) has provided new insights into winter–spring fatty acid dynamics. The observed patterns remain valid as they are linked to biological and ecological cycles that are consistent across years. A key limitation of this study was the use of only two seasonal sampling points. Future research should include year-round specimen collection to identify periods of peak and low n-3 fatty acid concentrations. This could inform decisions regarding their optimal utilization for human or animal consumption. This approach not only supports sustainable fisheries and circular economy principles but also promotes the valorisation of underutilized species, offering opportunities for income diversification and reduced reliance on traditional fish oil sources.

5. Conclusions

This study provides a comparative assessment of the fat content and the fatty acid composition in six Mediterranean fish species that are of low commercial value and often discarded when undersized. By combining univariate and multivariate analyses, consistent winter–spring differences in lipid profiles were identified, with spring-caught specimens generally showing higher fat levels and a greater proportion of nutritionally important n-3 LC-PUFAs such as EPA and DHA.

While the presence of these fatty acids is not in itself new, the novelty of this work is in the cross-species discard-focused approach and the integration of biochemical data with valorization potential for both aquafeed and processed human food products. This provides an evidence base for the sustainable use of underutilized biomass within the context of the EU Landing Obligation and circular economy strategies.

Future research should include year-round sampling to capture full annual variability and evaluate processing methods that enhance the safety, stability and marketability of these species for diverse end uses.