Decoding Fish Origins: How Metals and Metabolites Differentiate Wild, Farmed, and Escaped Specimens

1. Introduction

The Mediterranean Sea exhibits unique environmental characteristics due to its complex topography, geographical location between the subtropics and mid-latitudes, and dynamic meteorological patterns. These factors contribute to the frequent formation of low-pressure systems, as documented in several climatological studies [1]. The increasing prevalence of extreme warm events in the Mediterranean Basin aligns with global climate change trends and rising seasonal average temperatures. Multiple simulation models have predicted that such climatic shifts may lead to an increased frequency of explosive cyclones and a higher probability of hurricane formation in the region [1,2]. These environmental changes pose a significant threat to aquaculture, a sector reliant on stable marine conditions for the cultivation and production of aquatic species [3].

Globally, fish provide approximately 17% of animal protein and 7% of total protein consumed, supplying nearly 20% of the daily protein intake for over 3.2 billion people. The contribution of aquaculture to global fisheries has increased steadily, reaching 46.8% of total seafood production by 2016 [4]. The industry continues to expand at a rate surpassing that of other food production sectors. However, as aquaculture intensifies, traditional extensive farming systems are being replaced by high-density, intensive practices that increase the risk of disease outbreaks within cultures populations [5].

Beyond economic losses, extreme weather events such as storms can severely damage aquaculture infrastructure and lead to large-scale fish escapes. Such was the case with Storm Gloria in the winter of 2020, which caused significant structural damage to fish farms along the coasts of Valencia and Murcia, allowing large numbers of cultures fish to escape into the wild [6]. Escaped fish, which are typically bred in controlled environments and fed artificial diets, differ significantly from their wild counterparts in both behavior and physiology. Their introduction into natural ecosystems poses several ecological and public health risks, including competition with native fish for food and habitat, genetic pollution through interbreedin [5,7,8].

Izquierdo-Gomez et al. [9] categorize fish escapes into three levels: routine escapes (approximately 5,000 fish per year, or 1.31 tons annually), mass escape events (up to 91 times the normal level), and catastrophic events (up to 1,800 times the normal level). The escape of cultures fish can result in the establishment of self-sustaining populations, alter genetic diversity, introduce pathogens, and disrupt local food webs [10]. In the Mediterranean Sea, cultures fish escapes have been linked to genetic hybridization and competition with native species for resources [11]. These events can also impact commercial fisheries, as escaped fish often mix with wild stocks, complicating stock assessments and reducing market value. The ability to distinguish cultures fish from wild specimens based on morphological differences—such as a more rounded and compact body shape in cultures fish—can aid in assessing the impact of these escapes [12].

One of the main challenges in managing fish escapes is accurately identifying escaped individuals once they enter natural ecosystems. The ability to distinguish between wild, cultures, and escaped fish is critical for fisheries management, conservation efforts, and ensuring the authenticity of seafood products in markets.

The analysis of heavy metals and fatty acids has emerged as a valuable tool for fish traceability. Fish accumulate metals such as arsenic (As), selenium (Se), and mercury (Hg) from their environment and diet, and the concentrations of these elements can vary depending on whether the fish were raised in aquaculture facilities or in the wild [13,14]. Cultures fish are often exposed to different dietary formulations and water conditions compared to their wild counterparts, leading to distinctive metal accumulation patterns. For instance, cultures fish tend to have lower levels of Hg and higher levels of Zn and Cu due to the composition of commercial aquafeeds. Inductively Coupled Plasma Mass Spectrometry (ICP-MS) is one of the most advanced techniques for detecting and quantifying trace metals in fish tissue. This method offers high sensitivity and precision, allowing for the identification of even trace concentrations of elements. ICP-MS analysis is particularly valuable for seafood traceability, as differences in metal content can reflect variations in feeding regimes, aquaculture water sources, and environmental exposure [14,15].

Similarly, fatty acid profiles serve as robust biochemical markers for differentiating fish from different origins. Wild fish naturally acquire a diet rich in omega-3 fatty acids (e.g., eicosapentaenoic acid [EPA] and docosahexaenoic acid [DHA]), whereas cultures fish are typically fed diets that include plant-based oils, leading to higher levels of omega-6 fatty acids, particularly linoleic acid (18:2n-6). The presence of certain fatty acids, such as linoleic acid, in fish muscle tissue can thus be indicative of aquaculture origin [16]. The ratios of omega-3 to omega-6 fatty acids, along with specific lipid biomarkers, can effectively distinguish between wild fish, escaped fish, and those reared in aquaculture facilities [6]. Gas Chromatography-Mass Spectrometry (GC-MS) is widely used to analyze fatty acid profiles in fish muscle tissue. This technique involves the conversion of fatty acids into volatile methyl esters, which are then separated and quantified. GC-MS provides detailed insights into lipid composition, allowing for the identification of key fatty acids such as EPA, DHA, and linoleic acid. The ratio of omega-3 to omega-6 fatty acids serves as a reliable indicator of dietary origin, distinguishing between wild-caught and cultures fish [6].

Beyond heavy metal and fatty acid analysis, Metabolomics has emerged as a powerful approach for studying biochemical differences between wild, escaped, and cultures fish. Metabolomics involves the comprehensive analysis of small-molecule metabolites present in biological tissues, providing insights into metabolic pathways influenced by diet, environment, and physiological status.

Nuclear Magnetic Resonance (NMR) spectroscopy is a widely used analytical tool in metabolomics, allowing for the identification and quantification of both polar and apolar metabolites [17]. By applying NMR-based metabolomics, researchers can detect metabolic differences that arise due to variations in diet, environmental exposure, and physiological adaptation [18]. Cultures fish, for example, exhibit distinct metabolic signatures characterized by higher levels of creatine, creatinine, lactate, and alanine, reflecting the impact of artificial feeding regimes and controlled rearing conditions [6]. In contrast, wild fish demonstrate higher levels of taurine, an amino acid obtained from natural prey, and an increased abundance of omega-3 fatty acids.

The application of metabolomics to escaped fish is particularly relevant in assessing how their metabolic profiles change once they enter natural ecosystems. Escaped fish often experience abrupt shifts in diet and environmental conditions, leading to metabolic adaptations that can be detected through NMR-based metabolomic profiling. This approach provides a unique means of assessing fish traceability, complementing traditional chemical analyses [6].

The metabolomic signatures of wild and cultures fish differ significantly due to variations in diet and metabolic activity. Cultures fish typically exhibit higher levels of metabolites associated with artificial feeding regimes, such as: Creatine and creatinine, indicators of higher energy reserves due to regular feeding; lactate and alanine, metabolites linked to glycolysis and anaerobic metabolism, which are more prominent in cultures fish with lower physical activity; and glycine and betaine, amino acids commonly found in aquaculture feeds. In contrast, wild fish show higher levels of metabolites that reflect a more diverse and natural diet: TMAO, a key metabolite obtained from marine prey, found at significantly higher concentrations in wild fish; or omega-3 fatty acids (EPA, DHA), essential fatty acids derived from marine food sources, crucial for cell membrane integrity and metabolic function. In this line, the Lipidomics, a subset of metabolomics, focuses on the comprehensive analysis of lipid species, providing critical insights into dietary influences and metabolic adaptations. The fatty acid composition of fish muscle tissue is a particularly valuable marker for traceability. Wild fish exhibit a lipid profile rich in omega-3 fatty acids, with dominant peaks corresponding to EPA (20:5n-3) and DHA (22:6n-3), and cultures fish have a lipid profile characterized by higher levels of omega-6 fatty acids, particularly linoleic acid (18:2n-6), due to the inclusion of plant-based oils in aquafeeds [19]. Escaped fish could be present an intermediate profile, reflecting dietary changes as they transition from aquafeed consumption to foraging in the wild.

Given the increasing incidence of fish escapes and the need for reliable traceability methods, this study aims to integrate multiple analytical approaches—metals analysis, fatty acid profiling, and NMR-based metabolomics—to differentiate between wild, cultures, and escaped fish. The specific objectives include: Characterizing heavy metal concentrations in gilthead seabream, European seabass, and meagre to assess their potential as traceability markers; analyzing fatty acid profiles using GC-MS to distinguish fish origin based on lipid composition; applying NMR-based Metabolomics and Lipidomics to identify key biochemical biomarkers for fish traceability; and developing a classification model for determining the origin of fish with high accuracy.

This comprehensive approach will contribute to improved fisheries management, aquaculture sustainability, and the prevention of seafood fraud.

2. Materials and Methods

2.1. Specimen Collection and Sample Preparation

Fish samples, 30 gilthead seabream (Sparus aurata), 30 European seabass (Dicentrarchus labrax), and 20 meagre (Argyrosomus regius), were collected from fish markets, supermarkets, and local and wholesale markets in the Valencian Community and Murcia (Spain) between 2019 and 2022. Each specimen underwent an initial external examination to assess the presence of parasites, followed by biometric measurements and photographic documentation. Muscle tissue samples were subsequently extracted and stored at -20 °C until analysis. Based on their appearance, the presence of regenerated scales [9] and traceability references (‘commercial’ labelling), the fish were classified into the following origin groups: ‘Wild’, ‘Escape’ and ‘Cultured’.

2.2. Heavy Metals Analysis by ICP-MS

Metal profiles were obtained from the digestion of muscle samples and ICP-MS analysis. These analyses were carried out at the Technical Research Services (SSTTI) of the University of Alicante. Fifteen trace elements selected on the basis of their frequency, bioaccumulation capacity and potential detrimental effect on human health (Al, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Se, Mo, Cd, Tl, Pb, Hg) were quantified. For chemical analysis, muscle samples were taken from the same individuals used in the previous biochemical analyses, i.e., 10 specimens per species (Seabream, Seabass, Meagre) and group of origin (wild, escaped, cultured), except for wild seabass, for which sufficient specimens were unavailable during the escape event period.

The frozen samples were subjected to a digestion process in a solution of 4 ml of HNO₃ and 0.5 ml of H₂O₂. The program consisted of different phases, reaching a final temperature of 240 ºC over 45 min. Afterwards, the resulting sample was diluted to 15 ml with milli-Q water. Aliquots were taken and processed by standard ICP-MS analysis [20], extracting the concentration of the different heavy metals listed previously.

2.3. Fatty Acid Analysis by GC-MS

The fatty acid profiles were constructed by extracting a battery of 35 of the most common fatty acids (Annex) by direct extraction of fatty acid methyl esters (FAME) [21]. The analysis performed by the external Technical Services of the Institute of Animal Science and Technology - UPV, provided a quantitative description of the different fatty acids, as well as a chromatogram corresponding to each individual.

In the experimental process, muscle samples from the specimens used for metabolomic analysis were used. The experimental process consisted of direct FAME synthesis [21]. In detail, the frozen samples were divided into 0.5 g subsamples which were ground at room temperature between 10 and 15 seconds. The derivative was placed in Pyrex tubes with: 1 ml of C13:0 standard (0.5 mg C13:0/ml MeOH); 0.7 ml of 10 N KOH in water; and 5.3 ml of MeOH. The tubes were incubated in a water bath at 55 °C for 1.5 h with 5 sec shaking every 20 min to dissolve and hydrolyze the sample. Subsequently, the samples were brought to room temperature by applying a cold bath and 0.58 ml of 24 N H₂SO₄ in water was added. The tubes were mixed by inversion and incubated in water at 55 °C for an additional 1.5 h, shaking them gently for 5 s every 20 min. After a second cooling, 3 ml of hexane were added and homogenized in vortex for 5 min. Subsequently, they were centrifuged for an additional 5 min. The hexane layers, containing the FAMEs, were transferred to gas chromatography vials, stored at -20 °C. Gas chromatographic (GC) analysis followed a standard protocol [21]. The GC-MS results were the concentrations of each fatty acid in mg per 100 g of sample.

2.4. 1H NMR Acquisition and Data Processing Parameters

A 500 µL sample was place in a 5 mm NMR tube, and spectra were referenced to TSP at 0.00 ppm (polar samples) or to chloroform at 7.26 ppm. All 1H NMR experiments were performed on a Bruker Avance 400 MHz equipped with a 5 mm HBB13C TBI probe with an actively shielded Z-gradient. The 1D solution state 1H NMR experiments had a 2 s recycle delay, 32,768 time-domain points and 2.556 s acquisition time. In total, 1024 scans were performed, and the experiment was conducted at 298ºK. Spectra were apodised through multiplication with an exponential decay, producing a 0.3 Hz line broadening in the transformed spectrum. The 1H NMR spectra were normalised and reduced to ASCII files using TopSpin (Bruker) and aligned using icoshift (version 1.0; available at www.models.kvl.dk) [6]. Processing of 1H NMR spectra was performed in MATLAB (MathWorks, Natick, MA, USA). The region of water (4.60–4.95 ppm) and extreme high and low fields (<0.5 ppm and 10 ppm, respectively) were removed. Metabolites were identified in one-dimensional spectra using The Human Metabolome Database (HMDB, https://hmdb.ca/) and the literature cited in this study [6].

2.5. Statistical Analysis

Data from metal and fatty acids quantification were analyzed using Nonclassical Multidimensional Scaling (nMDS), specifically employing the Sammon mapping algorithm to visualize sample relationships based on pairwise similarity matrices. This method reduces dimensionality while preserving topological distances, emphasizing subtle group differences.

Metabolomic data (polar and apolar fractions) were organized into feature matrices and analyzed using Partial Least Squares–Linear Discriminant Analysis (PLS-LDA) in MATLAB [22]. Pareto scaling was applied, and three components were used to build classification models. Model performance was evaluated through standard statistical parameters including R²X (explained variance in predictors), R²Y (explained variance in response), sensitivity, specificity, and area under the ROC curve (AUC).

3. Results and Discussion

3.1. Heavy Metals Analysis in Seabream, Seabass, and Meagre

The present study evaluated the fatty acid profiles and heavy metal concentrations as potential tracers for distinguishing escaped fish from wild and cultured counterparts. The findings reveal species-specific variations in metal accumulation and fatty acid composition, highlighting their potential use in traceability and food safety considerations.

The analysis of heavy metals in fish muscle tissue revealed significant differences between wild, escaped, and cultures fish (Figure 1, Figure 2 and Figure 3). Using Inductively Coupled Plasma Mass Spectrometry (ICP-MS), we quantified the concentrations of 15 trace elements, including Al, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Se, Mo, Cd, Tl, Pb, and Hg. Wild seabream exhibited higher levels of arsenic (As), selenium (Se), and mercury (Hg) compared to farm-raised fish. This is attributed to their natural diet, which includes benthic organisms and filter-feeding mollusks known for bioaccumulating arsenic. Notably, arsenic in wild seabream is primarily present in the form of arsenobetaine, which is considered non-toxic to humans [23]. The presence of arsenic in seafood has been widely studied due to its potential health risks, particularly when present in inorganic forms, which are highly toxic. However, marine organisms tend to accumulate organic arsenic forms, which have lower toxicity. Despite this, continuous monitoring is required to assess the potential long-term effects of arsenic accumulation in marine food chains.

The metabolism of arsenic in aquatic organisms is a complex process influenced by environmental and biological factors. Arsenic exists in both inorganic (iAs) and organic (oAs) forms, with arsenobetaine (AsB) being the predominant non-toxic form in marine fish [23]. The biotransformation of arsenic involves redox reactions and methylation processes, which are mediated by microbial and enzymatic activity [24]. In aquatic environments, arsenic can induce oxidative stress by generating reactive oxygen species (ROS), which can damage cellular components such as lipids, proteins, and DNA [25]. Antioxidant enzymes, including superoxide dismutase (SOD) and glutathione peroxidase (GPx), play a crucial role in mitigating oxidative stress, highlighting the intricate balance between arsenic toxicity and cellular defense mechanisms [26].

Selenium is an essential micronutrient that plays a crucial role in antioxidant defense, immune function, and thyroid hormone metabolism. Marine fish generally contain moderate to high selenium levels, which contribute to their nutritional value. Interestingly, selenium has been shown to counteract mercury toxicity by forming biologically inactive Se–Hg complexes, which reduce mercury’s bioavailability and toxicity [27]. The protective role of selenium is particularly relevant in wild seabream, where higher Se levels may mitigate the adverse effects of elevated mercury concentrations. However, excessive selenium intake can also be detrimental, leading to toxicity symptoms such as oxidative stress and metabolic disturbances. The balance between selenium and mercury is, therefore, an important factor in seafood safety and should be further investigated in future studies.

Mercury levels were also elevated in wild seabream, likely due to bioaccumulation through the food chain, given that seabream is a carnivorous species. The trophic transfer of mercury from prey to predator results in higher concentrations in top consumers, which is a concern for both ecosystem health and human consumption. The observed differences in metal concentrations suggest that metal profiles can serve as biomarkers to differentiate wild seabream from cultured counterparts.

In contrast, seabass did not show significant differences in heavy metal content between wild and cultured groups. The similarity in metal accumulation suggests that cultured seabass diets may closely resemble those of wild seabass, leading to a homogenization of metal profiles. This observation limits the utility of heavy metals as a distinguishing factor for seabass traceability. Additionally, seabass are more pelagic feeders compared to seabream, potentially leading to a different bioaccumulation dynamic that reduces metal differentiation.

For meagre, which included only cultured and escaped groups, escaped specimens exhibited higher concentrations of As, Se, and Hg. These findings suggest that escaped fish consume wild prey, leading to a bioaccumulative effect similar to that observed in wild seabream. The absence of wild meagre in this region of the Mediterranean prevents direct comparisons, but the observed differences highlight the potential of heavy metal analysis in tracing fish escapes.

3.2. Fatty Acid Profiles in Seabream, Seabass, and Meagre

The fatty acid analysis yielded patterns consistent with heavy metal findings. In seabream, distinct lipid biomarkers enabled differentiation between wild, cultured, and escaped groups. Wild seabream displayed higher levels of arachidonic acid (C20:4n6) and docosapentaenoic acid (DHA; C22:5n-3) compared to cultured individuals. These differences are linked to dietary variations, as wild seabream consume marine-derived lipids rich in long-chain polyunsaturated fatty acids (PUFAs), while cultured fish are fed diets containing vegetable oils [28,29].

Fatty acids play fundamental roles in fish metabolism, influencing membrane fluidity, signaling pathways, and energy storage. DHA, for instance, is critical for neural development, vision, and immune function in fish. It is also a key component of phospholipids in cell membranes, enhancing membrane stability and function. Arachidonic acid (ARA) serves as a precursor for eicosanoids, which regulate inflammatory responses, immunity, and stress adaptation. Wild seabream’s elevated DHA and ARA levels reflect their natural diet, which provides a more balanced and diverse lipid profile compared to cultured fish [30,31,32].

Conversely, cultured seabream exhibited elevated levels of linoleic acid (C18:2n6c) and linolenic acid (C18:3n3), consistent with the use of terrestrial plant-derived feed ingredients [30]. While these fatty acids can be elongated and desaturated to form long-chain PUFAs, their conversion efficiency in marine fish is relatively low. The higher prevalence of plant-derived fatty acids in cultured fish has implications for nutritional quality, as these lipids may alter the functional properties of cellular membranes and reduce the availability of essential fatty acids for physiological processes [33].

Seabass and meagre did not show significant differences in fatty acid profiles between wild and cultured individuals, aligning with the findings from heavy metal analysis. The lack of differentiation suggests that fatty acid composition alone may not serve as a reliable biomarker for traceability in these species. However, in seabream, normalization of fatty acid values improved classification, supporting the use of lipidomics as a tool for distinguishing fish origin (Figure 4, Figure 5 and Figure 6).

3.3. Implications for Food Safety and Traceability

The findings of this study have important implications for seafood safety and traceability. The concentrations of heavy metals in all fish samples remained below regulatory safety limits, ensuring that consumption poses no immediate health risks. However, the bioaccumulation of arsenic and mercury in wild and escaped fish raises concerns about long-term exposure effects, particularly for consumers with high seafood intake. The selenium-to-mercury molar ratio observed in wild seabream suggests a mitigating effect on mercury toxicity, but further research is needed to determine the threshold for safe consumption [34].

From a traceability perspective, seabream demonstrated the greatest potential for differentiation based on both heavy metal and fatty acid profiles. The clear distinction between wild and cultured individuals highlights the feasibility of using these biomarkers in monitoring programs for escaped fish. In contrast, seabass and meagre require additional molecular or isotopic analyses to improve classification accuracy. The integration of fatty acid profiling with stable isotope analysis may enhance differentiation by providing insights into trophic interactions and feeding histories.

The data for metals and AAGGS analyses were analyzed using the MDS method. Variations in the concentrations of metals and AAGGs in the different types of fish (cultured, escaped or wild) can be visualized by Sammon mapping (Figure 7). Sammon mapping was used as this method diminishes the influence of large distances, which can completely dominate the map. The negative part of the x-axis in the Sammon mapping for metals was dominated by cultured and escaped fish samples, with wild fish samples dominating the positive part, due to the higher concentration in As, Se and Hg (Figure 1, Figure 2 and Figure 3). These samples are from sea bream. Sea bream are fish that consume algae that accumulate these metals. The wild sea bass samples, since they do not have these differences in these metals, are placed next to the cultured and escaped fish. For Sammon's map for AAGGs, we have a similar distribution, but here, on the negative side we do not have the wild sea bass samples, which are grouped with the wild sea bream samples. The AAGGs profiles are very similar in wild fish for both sea bream and sea bass (Figure 4, Figure 5 and Figure 6).

3.4. Metabolomic and Lipidomic Profiling Using NMR

This study aimed to evaluate the utility of metabolomic and lipidomic profiling in distinguishing between cultured, escaped, and wild fish specimens, specifically focusing on seabass (Dicentrarchus labrax) and meagre (Argyrosomus regius). The metabolomic and lipidomic analyses revealed significant differences between groups, with particular emphasis on the role of trimethylamine N-oxide (TMAO) as a key biomarker distinguishing wild fish from cultured and escaped fish [6,19,35,36,37].

The NMR-based metabolomic analysis of the polar fraction from seabass muscle revealed a clear separation between cultured and wild specimens, while escaped fish exhibited intermediate characteristics (Figure 8). The multivariate PLS-LDA analysis effectively grouped wild fish apart from the cultured and escaped specimens were made for seabream [6]. We have used the build model for sea bream and tried to classify sea bass and sea bass specimens. We obtained an excellent classification of the samples (Figure 9), suggesting a good tool for traceability.

A key finding was the significantly higher levels of TMAO in wild seabass compared to cultured and escaped specimens. TMAO plays a critical role in osmoregulation in marine fish, stabilizing proteins against denaturation and oxidative stress in varying salinity conditions [38,39]. Wild fish, exposed to fluctuating salinities, accumulate higher levels of TMAO, whereas cultured fish, kept under stable conditions, exhibit lower levels. Escaped fish had intermediate levels, suggesting partial metabolic adaptation but not full alignment with wild conditions. These findings are consistent with recent studies on gilthead seabream (Sparus aurata), which also reported significantly elevated TMAO in wild fish, indicating its role as a robust metabolic marker of environmental adaptation [37,40,41,42,43].

TMAO serves multiple physiological functions beyond osmoregulation. It has been identified as a key metabolite for stabilizing proteins and membranes against environmental stressors such as temperature fluctuations and oxidative stress. Studies indicate that TMAO enhances mitochondrial efficiency, contributing to improved energy metabolism in wild fish [41]. Additionally, wild fish tend to have a more diverse gut microbiota, which enhances TMAO biosynthesis. In contrast, cultured fish, often exposed to antibiotic treatments, exhibit reduced microbial diversity, limiting their ability to produce TMAO [38].

In aquaculture, lower TMAO levels might indicate reduced adaptation to natural environmental conditions and could be linked to inferior muscle quality. Furthermore, TMAO levels in wild fish have been correlated with increased muscle firmness, a key quality attribute for seafood consumers [40].

For meagre, a similar pattern emerged, with TMAO serving as the primary distinguishing metabolite. The intermediate TMAO levels in escaped meagre reinforce the idea that metabolic adaptation to the wild is an ongoing, time-dependent process [40].

Furthermore, cultured fish showed elevated levels of betaine, creatine, and glucose, reflecting metabolic adaptations associated with captivity. These metabolites, involved in energy storage and anaerobic metabolism, highlight the restricted mobility and high-energy diets typical of aquaculture environments [44,45,46]. The higher lactate levels in escaped fish suggest increased muscular activity, supporting the hypothesis of a metabolic shift following escape.

Lipid profiling using ¹H NMR on the apolar fraction further confirmed dietary influences on metabolic composition. Cultured fish exhibited significantly higher levels of linoleic acid (C18:2n6) and α-linolenic acid (C18:3n3), characteristic of vegetable oils used in aquafeeds. These fatty acids were absent in wild fish, underscoring diet as a key determinant of lipid composition (Figure 10). Wild fish showed higher concentrations of arachidonic acid (C20:4n6) and docosahexaenoic acid (DHA, C22:6n-3), derived from marine prey. These long-chain polyunsaturated fatty acids (LC-PUFAs) are essential for cell membrane function, neural development, and immune responses [47]. Their enrichment in wild fish reflects their natural diet, rich in marine-derived lipids, in contrast to cultured fish, which receive diets supplemented with plant oils.

The distinct triplet signal at 2.79 ppm, corresponding to di-unsaturated fatty acids (DUFAs), was exclusively found in cultured and escaped fish (Figure 10). This signal was linked to linoleic acid, confirming its role as a dietary biomarker for aquaculture feeds. Adjusting the classification model to include this biomarker improved the differentiation of wild fish, validating its potential use in traceability applications [6].

The fatty acid composition in fish muscle reflects not only dietary influences but also physiological adaptations to energy metabolism, immune function, and stress resilience. DHA, for instance, is a crucial component of neuronal and retinal membranes, playing a fundamental role in cognitive function and vision in fish [48,49]. Meanwhile, arachidonic acid (ARA) acts as a precursor for eicosanoids, which modulate inflammatory responses and immune defense.

Fish raised in aquaculture systems often have lower DHA and ARA levels due to their plant-based diets, which lack these essential lipids. This may have implications for immune competence, as studies have shown that fish with higher dietary ARA intake exhibit enhanced stress tolerance and disease resistance.

The PLS-LDA model created to classify the seabream specimens into farmed, escaped and wild according to the 1H NMR spectra of the apolar metabolites [6], was used to classify the spectra obtained from the seabass and meagre samples (Figure 9). In this case, as with the spectra obtained from the 1H NMR analysis of the polar metabolites, we also obtained an excellent separation and classification of the samples (Figure 11).

4. Conclusions

This study presents a multidisciplinary approach to trace the origin of fish through the integration of heavy metal analysis, fatty acid profiling, and 1H NMR-based metabolomics and lipidomics. By analyzing three economically relevant Mediterranean species—Sparus aurata (gilthead seabream), Dicentrarchus labrax (European seabass), and Argyrosomus regius (meagre)—we demonstrate that biochemical signatures can effectively distinguish between wild, farmed, and escaped individuals.

Our findings highlight the following key insights:

Heavy metals, particularly arsenic (As), selenium (Se), and mercury (Hg), were significantly elevated in wild and escaped gilthead seabream, supporting their potential as traceability biomarkers. These differences are linked to natural feeding habits and environmental exposure.

Fatty acid composition revealed species-specific patterns, with wild seabream showing higher levels of DHA and ARA, and cultured fish presenting elevated linoleic acid. These differences reflect dietary sources and feeding regimes, though they were less pronounced in seabass and meagre.

Metabolomic analysis, especially the quantification of TMAO, creatine, and betaine, provided robust discrimination between groups. TMAO emerged as a key marker of environmental adaptation and was consistently higher in wild fish. Escaped specimens displayed intermediate metabolic profiles, reflecting partial adaptation to natural conditions.

Lipidomic profiling confirmed the influence of aquaculture diets on muscle lipid composition. The presence of linoleic acid and other plant-derived fatty acids in cultured fish, and their absence in wild individuals, further reinforced dietary traceability.

The application of multivariate statistical models (MDS, PLS-LDA) enabled accurate classification of fish origin, particularly in gilthead seabream. These models may be adapted for broader use in seafood authentication programs.

Overall, this work demonstrates that combining metal, lipid, and metabolite analyses offers a reliable and integrative strategy for fish traceability. Such tools are essential for strengthening seafood authentication, detecting aquaculture escape events, supporting sustainable fisheries management, and protecting consumer trust.

Future perspectives should explore the integration of compound-specific isotope analysis, microbiota profiling, and long-term studies to enhance traceability systems. The approach presented here lays the groundwork for the development of molecular traceability platforms applicable to both regulatory and commercial contexts.

These findings contribute to improving seafood traceability, ensuring food safety, and supporting sustainable fisheries management.