1. Introduction
Aquatic products, including fish, shellfish, and aquatic plants, have played a pivotal role in culinary traditions and global food systems because of their nutrient profile and economic significance [
1,
2]. As the demand for protein and other nutritional components (e.g., omega-3 fatty acids) continues to surge with population growth, aquaculture’s steady supply of nutritious seafood is becoming crucial for food security and sustainable development [
3,
4]. Although practiced for over 2000 years, aquaculture has recently been transformed through emerging technologies in precise farming and productivity enhancement [
5-
8]. Recirculating aquaculture systems provide controlled environments for optimal fish growth and enable water conservation [
9,
10]. Big data and sensor technologies have been employed to monitor shrimp health, behavior, and growth, considerably enhancing farming productivity [
11].
With the rapid growth of the production and consumption of aquatic products, ensuring the safety of aquaculture environments and foods remains challenging. Aquaculture systems are vulnerable to contamination with marine toxins, heavy metals, microplastics, and pathogenic bacteria. These contaminants enter the food chain through industrial runoff, improper waste disposal, and bioaccumulation in aquatic organisms [
12]. Aquaculture products are at a greater risk of contamination than products from open seas (
Fig. 1) [
13]. Environmental pollution and improper handling along the supply chain pose significant risks to consumer health and industry sustainability [
14]. Some shellfish contain intrinsic biological toxins and
Vibrio marinopraesens, requiring careful handling [
15,
16]. Aquaculture environmental pollution approaches a tipping point as industrial wastewater containing contaminants enters aquaculture areas through rivers and runoff [
17]. These hazards accumulate in aquatic organisms through ingestion and the food chain [
18]. Additionally, aquatic products can be contaminated by microplastics during processing, storage, and packaging, potentially becoming breeding grounds for disease. Recent foodborne illness outbreaks from tainted products highlight the need for effective safety strategies.
Microplastics and heavy metals released from industrial and human activities enter rivers, lakes, and oceans, where they are accumulated in aquatic organisms and subsequently enter the human body through the food chain. Aquatic products can be contaminated with pathogens during farming, transportation, and processing, posing a risk to human health. Marine toxins produced by bacteria and harmful algal blooms accumulate in seafood such as shellfish and fish, which can cause food poisoning.
Monitoring and prevention of hazards such as marine toxins, heavy metals, microplastics, and pathogens from aquatic environments represent effective solutions to mitigate intoxication outbreaks [
19]. Currently, major shortcomings exist in the detection and removal of hazards in aquatic products [
20]. Most conventional detection methods for hazardous substances require complicated instrumentation and long-time tests, including inductively coupled plasma mass spectrometry (ICP-MS) and atomic absorption spectroscopy [
21,
22]. Marine toxin detection methods such as mouse-based bioassays and liquid chromatography with fluorescence detection have low specificity and high costs [
23]. For pathogen detection, the polymerase chain reaction (PCR) requires skilled operators [
24]. Microplastics are widely detected in aquatic products because these are ingested by aquatic organisms [
25,
26]. Conventional microscopic analysis cannot detect tiny plastic particles (e.g., nanoplastics) [
27], necessitating simple methods for detection and removal [
28,
29]. This is crucial for China’s maritime renaissance, affecting one-fifth of world’s population [
30], where over 80% of aquatic products come from aquaculture [
31].
Addressing global food safety challenges requires innovative approaches combining advanced technologies with sound regulatory frameworks. Nanotechnology has emerged as a promising frontier for enhancing hazard detection and separation in aquaculture environments and products [
32]. Nanostructures with unique physicochemical properties serve as a bridge between molecular- and nanoscale-hazards and macroscale sensors and absorbents [
33-
36]. Nanostructures have potential for timely detection, prevention, and removal of contaminants and pathogens from aquaculture environments. Their tiny sizes and large surface areas facilitate hazard recognition and binding. Gold nanoparticles (AuNPs) can be used as signal transducers for recognition events in optical and electrochemical sensing, enabling sensitive detection at low concentrations [
37-
40]. Metal-organic frameworks (MOFs) have large surface areas and good adsorption capacities, which make them excellent adsorbents [
41]. Surface modification provides nanostructures selectivity and stability against environmental factors, essential for practical applications [
42]. Biocompatible nanostructures are increasingly used in contaminant detection and separation to avoid environmental concerns. For instance, biodegradable zein nanoparticles encapsulating eugenol and garlic essential oils have been developed as antimicrobial agents to prevent fish diseases, replacing potentially toxic silver nanoparticles (AgNPs) [
43].
Advancements in the synthesis, characterization, and applications of nanostructures hold immense potential for safeguarding public health and for promoting the sustainability and competitiveness of the aquaculture and fisheries sectors. This review highlights the status of safety issues in aquatic products, revealing the multifaceted challenges persisting in the aquaculture industry. Subsequently, the collection of innovative investigations showcases the potential of nanostructure-based approaches for detecting and separating hazards from aquatic environments and products (
Fig. 2). The groundbreaking utilization of engineered nanostructures to detect and quantify some of the most common hazardous substances, including marine toxins, heavy metals, microplastics, and pathogens, suggests potential in safeguarding the last line of defense for aquatic product safety. This review highlights the potential of nanostructures to revolutionize the aquaculture industry and contribute to a safer, more sustainable future for global food production.
2. Marine toxins
Toxins in aquatic environments are produced by bacteria, cyanobacteria, and microalgae, with shellfish and finfish as vectors [
44]. The bioaccumulation of these marine toxins can cause widespread mortality among fish and shellfish, threatening human health through contaminated seafood consumption [
45]. Paralytic shellfish poisoning by saxitoxin (STX) causes acute symptoms within 30 min, potentially leading to cardiovascular shock or respiratory paralysis [
46]. The Codex Alimentarius Commission (CAC) recommends a limit of 800 μg STX equivalents per kilogram of shellfish flesh [
47]. Okadaic acid (OA) ingestion causes gastrointestinal symptoms, including vomiting, diarrhea, and abdominal pain [
48]. The European Union (EU) allows maximum contamination of 160 μg OA equivalents per kilogram of shellfish flesh (
Table 1) [
49]. Brevetoxin (BTX) causes neurotoxic poisoning with gastrointestinal, neurological, and cardiovascular symptoms [
50]. The CAC applies a threshold of 800 µg BTX-2 equivalents per kilogram of shellfish flesh (
Table 1) [
51]. Tetrodotoxin (TTX) causes dizziness and muscle weakness, with the European Food Safety Authority (EFSA) applying a threshold of 44 µg TTX equivalents per kilogram shellfish flesh, and a dose of 0.5-3.0 mg of TTX can even be lethal for an adult (
Table 1) [
52]. Overall, monitoring marine toxins originating from aquatic environments and food is crucial for safeguarding aquatic organisms and ensuring aquatic food safety (
Table 2 [
41,
48,
53-
56]).
Methods established to detect marine toxins include high-performance liquid chromatography [
57], liquid chromatography-tandem mass spectrometry [
58], and enzyme-linked immunosorbent assay (ELISA) [
59]. Although these methods show good sensitivity and accuracy, they require complex operation, expensive equipment, and long test times. Because of the need for seafood safety, rapid on-site determination is required, even with reduced reproducibility. Nanostructures enable simple and rapid toxin detection owing to their reactivity and signaling capability [
41]. Lai et al. [
53] developed a colorimetric immunoassay using AuNPs tagged with glucose oxidase (GOx) and anti-BTX B antibodies for BTX B detection, using magnetic beads (MBs) (
Fig. 3(a)). BTX B competed with a BTX-bovine serum albumin (BSA) conjugate that bridges the AuNPs tagged with anti-BTX antibody and GOx. After magnetic separation, unreacted complexes were introduced into an Ag(I)-3,3′,5,5′-tetramethylbenzidine (TMB) system. GOx catalyzed H
2O
2 production from glucose, converting SO
32- to SO
42-, releasing Ag
+ from Ag
2SO
3 to react with TMB, producing blue color. Higher BTX concentration resulted in fewer antibodies binding to BTX-BSA conjugates on MBs. The optical signal intensity was inversely proportional to BTX concentration. The immunoassay showed a 0.1 parts per trillion (ppt) detection limit, with test results correlating to a BTX ELISA kit.
The colorimetric assay approach has limitations. The human eye shows low sensitivity to subtle color changes [
60], and background color can interfere with detection accuracy. Fluorescence provides higher sensitivity and less interference with background color, which is suitable for toxin detection. Dou et al. [
41] combined zirconium-based nanoscale MOFs (NMOFs) with two different aptamers to develop fluorescence-based NMOF-aptasensors 1 and 2 for the detection of STX and TTX, respectively. In this study, two zirconium-based NMOFs were constructed with Zr6 clusters as metal nodes, using 4,4′-(1,2-diphenylethene-1,2-diyl)dibenzoic acid and 1,2,4,5-tetrakis(4-carboxyphenyl) benzene as their organic ligands, respectively. NMOF 1 and NMOF 2 emitted bright green and blue fluorescence originating from their ligands, respectively. The STX-aptamer and TTX-aptamer, tagged with tetramethylrhodamine (TAMRA), emitted red fluorescence under ultraviolet (UV)-visible light. When aptamers were anchored onto NMOFs, the fluorescence resonance energy transfer (FRET) was weak. Adding STX or TTX caused aptamers to bind targets and adopt specific configurations, enhancing the FRET effect (
Fig. 3(b)). The enhanced FRET effect produced visible color changes under 350 nm-light: NMOF-aptasensor 1 shifted from yellow-green to orange, and NMOF-aptasensor 2 from purple to orange. This strategy enabled the detection of the marine toxins STX and TTX with detection limits of 1.2 (0.35 parts per billion (ppb)) and 3.1 mol∙L
-1 (0.98 ppb), respectively). Testing in shellfish samples showed average recovery rates of 84.79% to 103.31%, with acceptable repeatability and precision. This research advances the combination of nucleic acid aptamers with nanostructures for toxin detection.
Aptamers such as single-stranded DNA or RNA have become popular recognition moieties owing to their stability, low cost, and comparable affinity with antibodies [
61,
62]. Combining aptamer recognition and nanostructures for signal transduction has created a platform for marine toxin detection. Zhao et al. [
54] developed a surface-enhanced Raman scattering (SERS)-based aptameric nano swab sensor for OA through enhanced plasmon coupling. By packing AuNPs on a paper swab via chemical reduction and depositing three dimensional succulent-like silver particles (3D-SS), they created an effective SERS substrate. The electromagnetic field between the Au layer and 3D-SS particles amplified the SERS signal. The AuNP assembly generates first-stage enhancement through interparticle coupling, whereas 3D-SS particles provide second-stage amplification through an expanded surface area and lightning rod effect. An aptamer against OA was equipped on the bilayer interface, forming a sandwich with Ag nano-octahedral-tagged complementary DNA labeled with sulfo-cyanine3 (Cy3), providing third-phase enhancement (
Fig. 3(c)). The aptamer complexes with OA and releases the tagged DNA in a turn-off assay. This multi-fold signal boost achieved sensitivity with a detection limit of 0.3 ng∙mL
-1. Testing OA in
Saxidomus purpuratus,
Mytilus edulis, and
Alectryonella plicatula yielded recovery rates of 94.2%-94.8%, 91.9%-106.4%, and 97.3%-104.3%, respectively, demonstrating the assay’s sensitivity and accuracy.
Ramalingam et al. [
48] designed an electrochemical microfluidic biosensor chip for OA detection. The system consisted of a polydimethylsiloxane (PDMS) microfluidic chip with a screen-printed carbon electrode (SPCE). The SPCE was modified by graphene-AuNP composite material and an OA-specific aptamer, using a ferro/ferricyanide redox probe for label-free electrochemical sensing. The interaction between the aptamer and OA caused changes in charge transfer, with higher OA concentrations decreasing the current output owing to the repulsion between the negatively charged redox probe [Fe(CN)
6]
3-/4- and aptamer sequence. However, the opposite occurred at lower concentrations (
Fig. 3(d)). This method enables OA quantification based on electrical signals. The detection limit was 8.0 pmol∙L
-1 (6.4 ppt), showing higher sensitivity than colorimetry, Raman spectroscopy, and fluorescence methods. Testing the OA concentration in mussels revealed a spiked recovery rate of 95.0%-104.1%, demonstrating reliable performance in complex food matrices. The electrochemical techniques enhanced detection reliability, whereas microfluidic biochips enabled convenient on-site testing.
Dual-mode detection strategy provides accurate signal readout for targeted analytes. Raza et al. [
55] developed a nanozyme-based dual-mode immunosensor for TTX detection using Fe(III)-tannic acid/Prussian blue (Fe-TA@PB) nanocomposite conjugated with goat anti-mouse immunoglobulin G (IgG). TTX-modified bovine serum albumin was immobilized on a microtiter plate as capture antigen. Increased TTX reduces IgG-conjugated Fe-TA@PB binding to immobilized TTX, causing reduced TMB oxidation and color change. The photothermal effect enables temperature-based detection using a smartphone infrared camera. TTX detection range is 0.1-100 ng∙mL
-1, with limits of 0.26 ng∙mL
-1 (colorimetric) and 0.44 ng∙mL
-1 (photothermometric). Nanozyme composites are also used for STX detection. Cho et al. [
56] developed an STX detection system using AuNP/Co
3O
4@Mg/ Al nanozyme and molecularly imprinted polymer (MIP). An STX-affinity peptide template enabled MIP-based indirect competitive detection. The system achieved a detection limit of 0.317 μg∙g
-1 and high STX recovery (90.7%-97.4%) in real samples.
Although conventional techniques remain essential for confirmatory marine toxin analysis, nanotechnology-based biosensors enable rapid, sensitive, and field-deployable toxin detection. Integrating molecular recognition elements (e.g., aptamers and antibodies) with nanostructures provides signal amplification through optical enhancement, energy transfer, and nanozyme catalysis, enabling sub-ppt to low-ppb detection.
3. Heavy metal
Heavy metals present in terrestrial and aquatic systems have led to water pollution, a pervasive environmental challenge globally. Unlike organic pollutants that degrade, heavy metals persist owing to their nondegradable nature, causing long-term ecological consequences [
63]. Mining, electroplating, smelting, chemical manufacturing, agriculture, and residential wastewater discharge contribute to heavy metal contamination of water resources and aquatic products [
64,
65]. Consumption of heavy metals through contaminated water or food can cause health issues, including neurotoxic effects, renal impairment, carcinogenic risks, and developmental delays [
66]. The detection and removal of heavy metals from water have gained significant attention. Nanomaterials have advanced heavy metal detection and separation [
67,
68]. Their adsorptive properties and reactivity, resulting from surface effects, small size effects, quantum effects, and macro quantum tunneling, enable heavy-metal-ion detection and extraction [
69]. New nanostructures have been developed for monitoring and removing heavy metals in water samples and aquatic products (
Table 3 [
70-
79]).
3.1. Detection of heavy metals
Current professional instrumentation such as atomic absorption spectroscopy and ICP-MS can analyze heavy metal species accurately, but they are expensive, time-consuming, and require skilled laboratory operation [
80]. For simple, low-cost detection for users without a professional background, nanostructures are increasingly used in heavy metal detection [
81]. He et al. [
70] developed a plasmonic silver film with SERS activity, called hydroxyoxime@Ag-polyvinylidene fluoride (HOX@Ag-PVDF), for ultra-sensitive Cu
2+ detection. An
in situ interfacial assembly method was utilized to modify the PVDF membrane with AgNPs and HOX molecules, which served as Raman reporters and target receptors. The oxygen and nitrogen atoms from the -OH and -C=N-OH groups, respectively, coordinate with Cu
2+, causing proton dissociation and forming a square planar copper complex (
Fig. 4(a)). The HOX@Ag-PVDF film showed increased intensity ratio between 1342 and 1322 cm
-1 with increasing Cu
2+ concentration. The method shows excellent selectivity for Cu
2+ detection, with a 52.0 pmol∙L
-1 detection limit. Cu
2+ recoveries in river and tap water samples ranged from 90.8% to 103.7%, indicating potential for real applications.
According to the nanomaterials database, noble metal nanoparticles (e.g., Au and Ag) are among the most commonly used nanomaterials [
82]. However, their high expense and disposable nature make them less preferred sensing materials for sustainability [
83]. Wang et al. [
71] developed recyclable nanoprobes for colorimetric detection of Hg
2+. DNA-modified AuNPs are prepared with T-T mismatches within double-stranded DNA to recognize Hg
2+, forming T-Hg
2+-T complex at mismatch sites [
84,
85]. T-T mismatch disrupts terminal base pairing, causing entropy-based repulsion and dispersing AuNPs, making the solution red. When Hg
2+ binds to T-T mismatched base pair, it transforms dispersed AuNPs into assemblies, redshifting the surface plasmon resonance peak and changing solution color from red to purple (
Fig. 4(b)). Adding isopropyl alcohol recovers individual DNA-AuNPs through double stranded DNA (dsDNA) denaturation, partially releasing Hg
2+. DNA-AuNPs can be recovered from nano-waste containing Hg
2+ with over 80% efficiency. The recovered DNA-AuNPs show improved sensitivity, lowering detection limit from 18.5 to 0.2 nmol∙L
-1. These probes successfully detect Hg
2+ in Atlantic salmon muscle tissue with recovery rates of 106%-164%, demonstrating reliability in analysis. This approach offers a strategy for recyclability and sustainable nanomaterials development.
However, the aforementioned methods are designed for single heavy metal detection. El-Desoky et al. [
72] developed an electrochemical sensor for concurrent detection of Pb
2+, Bi
3+, and Cu
2+. The sensor consisted of Fe
3O
4 nanoparticles and graphene nanosheets mixed in a ratio of 2% (w/w) Fe
3O
4 to 5% (w/w) graphene. Square-wave anodic stripping voltammetry quantified these heavy metal ions through electrochemical reduction and oxidation on the electrode surface, with subsequent current changes enabling identification and quantification (
Fig. 4(c)). Detection limits were 0.3 (0.06 ppb), 0.2 (0.05 ppb), and 0.9 nmol∙L
-1 (0.06 ppb) for Pb
2+, Bi
3+, and Cu
2+, respectively. In water sample tests, recoveries for Pb
2+, Bi
3+, and Cu
2+ were 75.2%, 109.4%, and 109.3%, respectively, which were obtained using the standard addition method. The simultaneous detection of three different heavy metal ions using the electrochemical sensor enables field detection of multiple metal contaminants.
To further improve real-time detection capability in Raman spectroscopy and electrochemical detection, Li et al. [
73] designed a novel smartphone-assisted fluorescence detection technique for swift identification of cadmium ions in oysters. The ratiometric probe comprised cadmium telluride (CdTe) quantum dots (QDs), silicon-oxide-coated copper nanoclusters (CuNCs@SiO
2), and 1,10-phenanthroline (Phen). When exposed to Cd
2+, the fluorescence emission of CdTe QDs increased owing to disrupted photo-induced hole transfer between QDs and Phen from Phen-Cd
2+ complex formation. Using a smartphone and Color Picker application to convert fluorescence signals to RGB values, Li et al. [
73] established a linear relationship between the green/red channel ratio and Cd
2+ levels (
Fig. 4(d)). Validation showed cadmium levels in oyster samples matched ICP-MS results, with concentrations in the ranges of 0.34-3.19 parts per million (ppm) (via smartphone detection) and 0.34-3.21 ppm (via ICP-MS). Aggregation-induced emission (AIE) nanomaterials enable fluorescence "turn-on" detection through restricted intramolecular motion. Hao et al. [
74] developed carbon dots (CDs) with AIE properties for smartphone detection of Cu
2+ and Hg
2+. Cu
2+ induces aggregation-enhanced fluorescence with 0.46 µmol∙L
-1 detection limit, whereas Hg
2+ causes static quenching via chemical complexation, resulting in a detection limit of 25.8 nmol∙L
-1. By providing real-time smartphone feedback, nanostructures demonstrate significant potential in intelligent heavy metal detection [
86-
88].
Methods involving nanostructures offer enhanced sensitivity, selectivity, and portability compared with tranditional spectrometric and electrochemical methods. Current strategies rely on diverse mechanisms, including plasmonic detection using colorimetry or SERS, electrochemical sensing, and smartphone-assisted fluorescence quantification. Next-generation nano-sensors for heavy metals should be a combination of laboratory-grade accuracy and field-level practicality in aquaculture safety monitoring.
3.2. Removal of heavy metals
Conventional strategies for heavy metal contaminant removal include chemical precipitation, ion exchange, filtration membranes, and electrocoagulation [
89]. However, these methods have drawbacks. Chemical precipitation uses many chemical reagents that may harm the environment or health. Filtration membranes effectively eliminate heavy metal ions but are expensive [
90]. Electrocoagulation involves challenges such as electrode passivation, uneven anode dissolution, and inconsistent coagulant production during long-term operation [
91]. Adsorption using nanomaterials is widely used for pollutant removal [
92]. Their large surface areas provide abundant active sites for heavy-metal-ion adsorption. Yap et al. [
78] developed a bio-sponge comprising an alginate network encapsulating reduced graphene oxide (GO) modified with iron oxide nanoparticles and attached multithiol pentaerythritol tetrakis (mercaptopropionate) molecules using photoinitiated thiol-ene click chemistry (
Fig. 5(a)). Kinetic studies show that the adsorbent follows a pseudo-second-order model for Cd
2+ and Pb
2+ ions (coefficient of determination (
R2) > 0.99), with electrostatic attraction as the main driving force. Upon deprotonation in an aqueous environment, the surface functional groups on the thiolated graphene bio-sponge (such as carboxyl, thiol, and oxygen-containing groups) become negatively charged, attracting positively charged Pb and Cd ions from water. The thiolated graphene bio-sponge showed high adsorption efficiency in milli-Q water (98.8% Pb
2+ and 90.2% Cd
2+); however, its adsorption efficiency decreased in seawater (63.1% Pb
2+ and 36.1% Cd
2+) because of blockage of adsorbent pores by organic matter. This result indicates the necessity of developing heavy-metal-adsorbing marine-specific materials.
Rubin Pedrazzo et al. [
79] synthesized an eco-friendly nanosponge with dense carboxyl groups through polycondensation between citric acid and line caps (LCs), a pea starch derivative (
Fig. 5(b)). The adsorption of cationic ions by nanosponge relies on negative charges in the polymer structure, forming complexes with metals. When introduced to artificial seawater with 50 ppm Cu
2+, the nanosponges removed 80%-84% of Cu
2+, achieving an adsorption capacity of 49-52 mg Cu
2+ per gram. Although effective, the material required a 24 h treatment to achieve the maximum adsorption efficiency. To shorten adsorption time and improve efficiency, Safari et al. [
75] developed a core-shell magnetic selenium nanocomposite (Fe
3O
4@SiO
2@Se) for Hg
2+ removal from water samples. The negatively charged selenium nanoparticles attract Hg
2+ through electrostatic interactions (
Fig. 5(c)), forming HgSe to extract Hg
2+ from seawater. Hg
2+ adsorption follows the Langmuir isotherm model (
R2 > 0.997) with 70.4 mg∙g
-1 maximum capacity and 0.0591 L∙mg
-1 Langmuir constant. Fe
3O
4@SiO
2@Se removes 94.28% of Hg
2+ in seawater within 20 min, demonstrating potential for efficient heavy metal removal.
Organic mercury species are known for their higher toxicity compared to Hg
2+ [
93]. Ma et al. [
76] developed a nanosorbent for adsorbing both organic and inorganic mercury, combining polyvinyl alcohol (PVA)-based aerogel with MoS
2 nanoflowers (MoS
2NFs) (
Fig. 5(d)). The adsorbent removes Hg
2+ through chelate formation between mercury and sulfur and ion exchange with oxygen-containing groups, while eliminating MeHg through hydrophobic interactions. The PVA-5/75MoS
2NF adsorbent maintained an optimal solid-liquid partition coefficient (
Kd) of 10
7 at pH values 3 and 10, remaining stable in simulated polluted seawater. For high MeHg concentrations (100 ppb), the adsorbent achieved 90% removal efficiency. This nano-adsorbent enables simultaneous adsorption of organic and inorganic mercury, reducing aquatic heavy metal pollution. The discharge of radioactive wastewater in Japan concerns aquacultural industries. Cheng et al. [
77] developed bacterial cellulose particles with nanoscale zero-valent iron (BCP-nZVI) for Co
2+ and Sr
2+ radionuclide adsorption, achieving 92.5% and 58.6% removal in seawater through redox, hydroxyl complexation mechanisms.
Removal of toxin metal ions (e.g., Cd2+, Pb2+, Cu2+, and Hg2+) using nanostructures involves target-specific capture and separation through magnetic separation, centrifugation, floating, and membrane filtration, based on chemical complexation and electrostatic attraction with functional ligands (e.g., carboxyl- and surlfur-containing groups).
4. Microplastics
Plastic products have led to severe plastic fragment pollution, now recognized as a major threat to organisms and human health [
94,
95]. These fragments are classified as macroplastics (>5000 μm), microplastics (1‒5000 μm), and nanoplastics [
96]. Plastic pollution in aquatic ecosystems impacts the aquaculture environment, products, and fisheries development [
97,
98]. The United Nations Environment Programme (UNEP) reported that 75 million to 199 million tonnes of plastic waste exists in oceans, comprising 85% of marine debris, with projections showing annual input reaching 23 million to 37 million tonnes by 2040 [
99]. In response, nanostructures have been developed for detecting and removing plastic fragments in aquaculture environments and products (
Table 4 [
100-
108]).
4.1. Detection of microplastics
Methods for detecting microplastics include visual inspection [
109], fourier transform infrared (FTIR) spectroscopy [
110], pyrolysis gas chromatography mass spectrometry (Py-GC/MS) [
111], and fluorescence staining [
112]. Visual inspection only detects larger microplastics (> 500 µm) and shows variability owing to observer bias. FTIR spectroscopy analyzes single microplastic particles (> 10 µm) per sample, requires water-free samples, and is labor-intensive. Py-GC/MS destroys samples during analysis and has low efficiency. Fluorescence staining is affected by dye concentration, solvent type, and temperature, with poor reproducibility [
113]. Nanostructure-based assays provide efficient monitoring of different-sized microplastics across environments. Lai et al. [
100] developed a method using AuNPs to label common microplastics (polyethylene (PE), polyvinyl chloride (PVC), polypropylene (PP), polybutylene terephthalate (PBT), polystyrene (PS), and polymethyl methacrylate (PMMA)) of sizes 50-1200 nm in environmental water, quantifying them with single-particle ICP-MS (sp-ICP-MS) (
Fig. 6(a)). The method combines acid digestion and cloud-point extraction, followed by AuNP labeling. It effectively detects microplastics in various water types (deionized water, drinking water, river water, lake water, outlet water of WWTP, and seawater), with recoveries of 72.9% to 92.8%. To minimize microplastics loss during sample pretreatment and improve analytical accuracy, Yang et al. [
101] developed a method for analyzing nanoplastics (PS, 50-1000 nm) in water using membrane filtration and SERS (
Fig. 6(b)). Silver nanowires (AgNWs) were self-assembled to form a filter membrane to concentrate nanoplastics, and detection was achieved using SERS activity to enhance the Raman signal. The membrane, self-assembled from AgNW of 60 nm diameter, combined physical interception and SERS activity. After filtering water samples, PS nanoplastics were captured, excited
in situ, and detected with a 785 nm laser. The method solves nanoplastic separation and preconcentration issues, avoiding sample loss from transfer steps. High retention rates (> 86.7%) and detection sensitivity (0.1 ng∙mL
-1) enable analysis of low concentration nanoplastics in aquatic environments.
Although SERS is widely used in environmental monitoring, uneven hot spot distribution causing detection inconsistency remains challenging. Inspired by dragonfly wings, Zhu et al. [
102] developed a method for detecting nanoplastics (PS, 800 nm) using Ag/ZnO nanorod arrays on PDMS films as a photo-induced enhanced Raman spectroscopy (PIERS) substrate (
Fig. 6(c)). The substrate adsorbs nanoplastics through hydrophobic interaction, with its periodic array improving detection reproducibility. UV light irradiation increases electron density between noble metal nanoparticles, enhancing SERS signal sensitivity. Detection limits in tap water, lake water, river water, and seawater were 25, 28, 35, and 60 μg∙mL
-1, respectively, with recoveries from 94.8% to 102.4%.
Awada et al. [
105] designed a fluorescencent probe using fluorescein-labeled hyaluronic-acid-conjugated polymer nanoparticles (F-HA-CPNs) for detecting microplastics (PP, PS, polycarbonate (PC), low-density polyethylene (LDPE), and high-density polyethylene (HDPE)) (
Fig. 6(d)). The nanoprobes form spherical particles of 194 nm diameter, with cores of diketopyrrolopyrrole-based polymer and surfaces functionalized with fluorescein-labeled hyaluronic acid. The probe binds to microplastics’ carbon backbone through van der Waals forces and hydrophobic interactions, achieving picomolar-level affinity (
Kd = 3.4-14 pmol∙L
-1), 10
5 times higher than free hyaluronic acid. Overall, the above nanostructure-enabled assays have demonstrated innovation for sensitive detection of microplastics in complex aquatic matrices, based on either the chemical labelling of microplastics or their intrinsic chemical structures.
4.2. Elimination of microplastics
Microplastic removal methods include membrane filtration [
114], flocculation techniques [
115], and biological removal [
116]. However, membrane filtration requires regular filter maintenance; thus, it is costly. Flocculation techniques are affected by sample turbidity and flocculant ionic strength, risking chemical pollution. Biological methods have low efficiency and depend heavily on microorganism types and environmental conditions [
117,
118]. Nanomaterials with large specific surface areas effectively remove microplastics, offering potential solutions for aquatic contamination [
119]. Grbic et al. [
106] developed a method using hydrophobic iron nanoparticles to extract microplastics from water (
Fig. 7(a)). The synthesized iron nanoparticles, modified with silane, adsorbed microplastics through hydrophobic interactions, achieving recovery rates of 93% (>1 mm), 81% (200 μm to 1 mm), and 92% (<20 μm) for microplastics of different sizes in environmental samples. However, the hydrophobic interactions of the nanoparticles with the microplastics are significantly weakened in the presence of lipophilic substances or biota (e.g., fat from fish tissue) in the sample. Chen et al. [
107] fabricated a zirconium MOF (UiO-66-OH) that removes various nanoplastics from seawater through electrostatic, hydrogen bonding, and van der Waals interactions (
Fig. 7(b)). The material achieved 95.5% removal efficiency and maintained 90.7% efficiency over ten cycles.
MOFs are promising adsorbents owing to their stability, large surface areas, microporous structures, and tunable properties. However, common MOF materials (e.g., UiO-66) require high temperature and pressure conditions with organic solvents. Pasanen et al. [
103] synthesized a magnetic zeolite imidazolate framework (nano-Fe@ZIF-8) using
n-butylamine and Fe
2+ in aqueous solution at room temperature to remove PS microplastics from tap water (
Fig. 7(c)). The hydrophobic ZIF-8 aggregates with PS microplastics through van der Waals forces. Its microporous structure (surface area: 1038 m
2∙g
-1) and small size ((50 ± 30) nm) enhance microplastic adsorption. The nano-Fe@ZIF-8 material can be magnetically recovered and removed ≥93.8% of polystyrene microspheres through hydrophobic interactions. However, magnetic nanomaterials and MOFs can release metal ions, posing ecological hazards, and are costly. Li et al. [
108] developed nano oil-in-water (O-W) emulsions using olive oil to remove PS and PMMA microplastics from seawater, tea leaves, and toothpaste supernatant (
Fig. 7(d)). The emulsions, used to remove microplastics from seawater, tea leaves, and toothpast supernatant, achieved removal rates of 75%, 80%, and 82%, respectively, through hydrophobic interactions, offering sustainable and environmentally friendly advantages of sustainability, environmental friendliness, and generalityfor microplastic pollution mitigation.
The degradation and chemical transformation of microplastics are crucial for sustainable development goals. Miao et al. [
104] developed an electro-Fenton-like (EF-like) approach using a TiO
2/graphite (TiO
2/C) cathode to degrade PVC microplastics (100‒200 μm) in deionized water (
Fig. 7(e)). The approach decomposed PVC through cathodic reductive dechlorination and hydroxyl radical oxidation. PVC underwent dechlorination by acquiring electrons from the TiO
2/C cathode, followed by hydroxyl radical oxidation to produce oxygenated organic intermediates, which were then mineralized to carbon dioxide and water. The dechlorination efficiency and removal rate for PVC plastics were 75% and 56%, respectively. This result indicates that the method is a viable and environmentally friendly strategy for microplastic degradation and transformation.
Microplastic removal can be achieved through nanostructure-based surface adsorption via various interactions (e.g., van der Waals forces and electrostatic and hydrophobic interactions) through physical filtration, magnetic separation, and collection. Additionally, degradation of microplastics relies on electrochemical or hydroxyl radical oxidation enabled by nanostructures.
5. Pathogenic bacteria
Pathogenic bacteria are a major threat to China’s aquaculture industry, causing infectious diseases in economically important species such as shrimp, tilapia, and large yellow croaker [
120]. Bacterial infections originating from
Vibrio spp.,
Aeromonas hydrophila (
A. hydrophila), and
Edwardsiella tarda (
E. tarda) cause mass mortalities and reduced feed conversion efficiency [
121]. Antibiotic overuse has raised concerns about antimicrobial resistance and environmental contamination [
122]. Foodborne pathogens such as
Vibrio parahaemolyticus (
V. parahaemolyticus) pose serious health and economic burdens globally [
123]. The World Health Organization (WHO) reports 600 million annual foodborne illness cases, causing losses of 110 billion [
124,
125]. Rapid pathogen quantification in high-risk aquatic products is essential [
126]. Nanomaterials development has created new methods for detecting and removing pathogenic bacteria in aquatic organisms and products (
Table 5 [
43,
127-
131]).
5.1. Detection of pathogenic bacteria
PCR and matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS) are currently the main methods for detecting pathogenic bacteria [
132]. Although PCR and MALDI-TOF MS are highly reliable, these methods require complex preprocessing and analysis, which make them time-consuming [
133]. Nanomaterial-based detection methods offer easier operation, high sensitivity, and faster testing for pathogen detection in aquatic products. Wu et al. [
127] developed an efficient single-step method to detect
V. parahaemolyticus in seafood using adapter-assisted SERS (
Fig. 8(a)). The method uses cysteamine-modified gold-coated PDMS (Au-PDMS) film as SERS substrate, with 4-mercaptobenzoic acid (4-MBA) and aptamer-modified AuNPs as signal probes, assembled through electrostatic interactions. When
V. parahaemolyticus is present, the aptamer binds to bacterial surface protein, causing conformational change. The signal probe dissociates from the membrane surface, decreasing Raman intensity of 4-MBA at 1592 cm
-1. Using PDMS membrane transparency and mechanical stability, this method combines aptamer affinity and the dual AuNP "hotspot" effect to achieve a wide linear range (1.2 × 10
2-1.2 × 10
6 colony forming unit (CFU)∙mL
-1) and a low detection limit (12 CFU∙mL
-1).
However, this method requires a large Raman spectrometer and prevents on-site detection. Xu et al. [
129] developed a portable fluorescent biosensor with nanoparticles-based immunomagnetic separation for detecting bacterial pathogens in shrimps (
Fig. 8(b)). The detection mechanism is based on the use of immunomagnetic nanobeads (immuno-MNBs) to specifically capture pathogenic bacteria and the use of CdSe/ZnS core/shell QDs modified with antibodies to label them. Consequently, an immuno-MNB-bacterial-QD "sandwich" complex is formed. The fluorescence intensity is measured by a portable spectrometer, with the process completed within 60 min. Detection limits for
Escherichia coli (
E. coli) O157:H7,
Listeria monocytogenes (
L. monocytogenes), and
Salmonella typhimurium (
S. typhimurium) in shrimp were 10
2, 10
3, and 10
3 CFU∙mL
-1, respectively, with recoveries of 89%-111%. Wang et al. [
130] developed an electrochemical aptamer sensor for rapid detection of
V. parahaemolyticus in shrimps (
Fig. 8(c)). The assay uses aptamer-modified magnetic nanometal organic frameworks as capture probes and AuNPs conjugated with ferrocene and phenylboronic acid as nano labels. The sandwich-type complex formed is magnetically adsorbed onto a screen-printed electrode (SPE), converting to an electrical signal. This detector recognizes
V. parahaemolyticus at 10-10
8 CFU∙mL
-1 within 20 min, with a 3 CFU∙mL
-1 detection limit and 94%-107% recovery.
Compared with SERS, immunofluorescence bioassay, and electrochemical detection, colorimetric analysis enables quantitative on-site detection of pathogenic bacteria without instruments, offering easy operation and low cost. Fu et al. [
131] developed a colorimetric method for detecting
V. parahaemolyticus in oysters using immunomagnetic separation and Mn
2+ ion-mediated AuNP aggregation (
Fig. 8(d)). When
V. parahaemolyticus is present, chicken egg yolk antibodies (IgY)-MBs and IgG-MnO
2 NPs bind to different sites of the target through antigen-antibody reactions, forming sandwich-type immune complexes. Unbound IgG-MnO
2 NPs are removed magnetically. Adding ascorbic acid etches MnO
2 to produce Mn
2+ ions proportional to bacterial concentration. These ions interact with AuNPs’ carboxyl groups, causing aggregation and color change from red to blue. This method detects
V. parahaemolyticus at 10-10
6 CFU∙mL
-1 with 10 CFU∙mL
-1 detection limit and 95.78%-101.83% recoveries, enabling on-site visual inspection. Combining nucleic acid amplification and AuNP indicators provides another colorimetric detection method [
134]. Liu and Wang [
135] integrated DNA-AuNPs with loop-mediated isothermal amplification (LAMP) assay for
V. parahaemolyticus detection. This naked-eye LAMP assay offers three advantages over PCR: specifically designed DNA sequences against the target gene enhance detection accuracy, completion within 50 min, and simple isothermal amplification without PCR instruments. These detection schemes use nanostructures as indicators and signal amplifiers coupled with aptamers, antibodies, and target gene-specific DNA sequences, facilitating selection based on accuracy, time, and cost requirements.
5.2. Inhibition of pathogenic bacteria
Antibiotics and other biocides effectively kill pathogens but require continuous use and cause drug-resistance concerns [
136]. Because of their high surface-area-to-volume ratio for bacterial adsorption, nanostructures are gaining research attention [
137]. Zhan et al. [
128] removed pathogenic bacteria from water using magnetic Fe
3O
4/graphene (G-Fe
3O
4) composites (
Fig. 9(a)). Graphene’s porous structure adsorbs and kills bacteria by disrupting cell walls, whereas Fe
3O
4's magnetic properties enable material separation and recycling. The G-Fe
3O
4 composite removes 94.8% of pathogenic bacteria, including
E. coli,
Staphylococcus aureus (
S. aureus),
Salmonella, and
Enterococcus faecalis (
E. faecalis), from river water. Protein nanoparticles, being green and sustainable, avoid toxic emissions that threaten health and environment. Luis et al. [
43] developed zein nanoparticles with eugenol and garlic essential oil to treat fish pathogens (
Fig. 9(b)). These 150 nm positively charged particles work through eugenol that destroys bacterial membranes and garlic’s allicin that inhibits enzymes. This combination enhances inhibition of
A. hydrophila,
E. tarda, and
Streptococcus iniae (
S. iniae). The nanocarrier enables sustained release, maintaining >82% active ingredients for 90 days and reducing nontarget toxicity with increased lethal concentration 50 (LC
50). This approach combining plant compounds with nanotechnology advances fish disease treatment and sustainable fisheries.
The aforementioned antibacterial approaches mostly involve nonspecific bacteria-killing chemicals and nanostructures. For specific inhibition of pathogenic bacteria, Zhang et al. [
138] developed a gold nanocluster (AuNC)-based nucleic acid delivery approach for bacterial gene suppression. The AuNCs enable efficient internalization of target gene-specific antisense oligonucleotides (ASOs) into Gram-positive and Gram-negative bacteria. The ASOs are released from AuNCs within bacteria, enabling ∼70% knockdown of
mecA in methicillin-resistant
S. aureus, reducing antibiotic resistance and enhancing oxacillin treatment. They also used CDs to deliver
empA-specific ASO to
Vibrio anguillarum (
V. anguillarum) for targeted EmpA suppression [
139]. An adhesive hydrogel boosted ASO/CD concentrations at wound sites in seawater, preventing fin rot disease in turbot and offering solutions for underwater bacterial diseases.
6. Conclusions and perspectives
Aquaculture, practiced for over 2000 years, is now recognized as one of the most promising food production systems worldwide and is integral to human society’s sustenance. Recently, wastewater discharge from industrial activities and climate change have severely impacted the aquaculture environment and aquatic product safety. Monitoring and reducing hazardous substances in real-time during aquatic product production is an urgent problem requiring solutions. Nanostructure-based detection and elimination of various hazards offer efficient routes to isolate and remove harmful species from aquatic environments and products, thereby mitigating risks to consumer health and enhancing safety of aquatic organisms and foods. This paper provides a systematic overview of the applications of advanced nanostructures (e.g., AuNPs, AgNPs, MOFs, and magnetic nanoparticles) in detecting and eliminating common pollutants, such as marine toxins, heavy metals, microplastics, and pathogenic bacteria. These nanostructures improve pollutant detection efficiency while minimizing large-scale equipment use, lowering technical and cost barriers for public use, and enabling real-time monitoring. Although significant progress has been achieved in this field, research opportunities and challenges remain ahead.
(1) Enhancement of stability of nanostructures. High chemical and colloidal stability of nanostructures is crucial for their performance in detection and separation applications. Currently, nanostructures face aggregation or degradation issues in complex environments. Light irradiation, temperature, pH fluctuation, salinity and organic matter often lead to functionality decay and application failure. In aquaculture, nanostructure properties are affected by water salinity, pH and temperature. For example, antibodies attached to AuNPs through electrostatic interactions for biomolecular sensing [
140] may be disrupted under different pH or salt conditions, compromising detection capability. Future nanostructure design should focus on stabilization in complex environments. Hydrophilic surface functionalization and charge optimization can maintain colloidal stability against contaminants, whereas chemical stability of reactive nanoparticles (e.g., AgNPs) can be improved by coating with inert compositions, such as SiO
2 and polyaniline [
141,
142]. Polymeric coating with polyethylene glycol, PVA, or dextran creates a hydrophilic layer preventing agglomeration [
143-
145] and improves chemical stability by preventing oxidation [
146].
(2) Enhancement of the detection sensitivity of nanostructures. Boosting nanoprobe sensitivity is essential for early detection of aquaculture hazards. Several strategies exist for sensitivity enhancement. Magnetic trapping and electrophoretic methods help preconcentrate analytes to lower detection limits. Signal amplification can be achieved through enzymatic amplification or nanoparticle enhancement. Efforts to optimize receptor molecular structures improve the target’s binding affinity. For example, the systematic evolution of ligands by exponential enrichment enables selection of high-affinity aptamers [
147]. Locked nucleic acids enhance aptamer binding affinity in aquaculture water [
148]. Engineering aptamers with multiple binding sites increases target affinity [
149]. Selecting aptamers within complex matrices optimizes recognition and eliminates sample purification requirement [
135].
(3) Enhancement of the multifunctionality of nanostructures. Integration of different functional properties into nanostructures enables multiple tasks, such as detection, capture, and removal of contaminants. Magnetic plasmonic nanocomposites containing magnetic and plasmonic nanoparticles allow optical detection and magnetic separation [
150]. Plasmonic nanoparticles decorated with ZnO or TiO
2 enable detection and photocatalytic degradation of toxins and microplastics, and pathogen elimination. MOFs with recognition moieties and nano-catalysts can detect, adsorb, and eliminate pollutants [
151]. Molecular imprinting polymers with magnetic nanoparticles enable recognition and separation [
152]. Multimodal sensors combining sensing modalities (e.g., colorimetry, fluorescence, SERS, and electrochemical sensing) can provide accurate detection capabilities [
153,
154]. Different functional nanoparticles coupled with distinct capture ligands enable multiplex detection and separation of various analytes. Responsive nanoparticles react to pH and temperature changes to facilitate contaminant detection and separation [
155]. Microfluidic devices with nanomaterials serve as platforms for real-time detection and elimination of hazards [
156,
157].
(4) Production and application of sustainable nanostructures. Sustainable nanomaterials for detecting and eliminating contaminants from aquaculture environments can minimize environmental impact while maximizing effectiveness. AuNP-based detection nanoprobes, CDs, and nanostructures made from bioresources are ideal for commercial uses, owing to their synthesis ease, low cost, recyclability, and environmental compatibility. Noble metal nanoprobes with DNA can be produced and recycled for multiple applications [
71,
135]. Producing nanomaterials from waste products such as carbon- and silica-based nanostructures contributes to circular economy because of the conversion of waste into valuable resources. Industrial byproducts may be repurposed into nanomaterials. Biodegradable biopolymers, such as chitosan, alginate, cellulose, and polylactic acid, can be investigated as precursors for functional nanomaterials [
158,
159]. Green synthesis using bacteria or plant extracts as reducing agents should be developed for nanomaterial preparation [
160,
161]. The cost for treating 1 tonne of seawater containing heavy metals using high-carboxyl nanosponge is approximately 3000 CNY [
78]; microplastics removal using Fe nanoparticles costs ∼1500 CNY [
105]; pathogen elimination using magnetic G-Fe
3O
4 costs ∼2700 CNY [
122]. Despite high costs, these nanomaterials provide excellent elimination performance without environmental harm. A sustainable supply chain should be developed with industrial partners, focusing on responsible sourcing and green chemistry. Stimuli-responsive nanomaterials have emerged as sustainable options, capturing targets under specific conditions and releasing them upon environmental changes [
162,
163]. A lifecycle assessment is needed to evaluate nanomaterial environmental impact.
(5) Development of standardized protocols and integration of artificial intelligence (AI) into the design and application of nanostructures. AI algorithms such as deep learning can predict and design new nanomaterials for contaminant detection and elimination [
164]. The functionality of nanomaterials can be optimized using AI through size, shape, and surface chemistry. For hazard detection applications, standardized methods for field testing using nanostructures are essential in aquaculture. Standardized devices should be developed for pretreating seawater, aquaculture water, and aquatic product samples to ensure consistent processing into comparable solutions for toxin detection. This approach would enhance result reliability and facilitate standardized methods for toxin detection. Once standardized protocols are established, integrating nanostructures-based testing into routine monitoring systems will improve hazard detection, enhanced by AI integration. Machine learning algorithms can be developed for sensor data analysis using spectral or image datasets. AI integration with spectroscopic techniques (e.g., Raman, FTIR, and UV-visible spectroscopy methods) can enhance hazard detection by training models to interpret spectral data and identify contaminants. Image datasets of microplastics, pathogens, and contaminants can train AI algorithms for automated identification based on physical properties, improving detection accuracy and reducing false results.
(6) Establishing standards and regulations for applying nanostructures in aquaculture. The aquaculture industry must address biosecurity, aquatic organism health, food safety, and environmental health. The use of nanostructures should align with established national and local standards. Standards and regulations must be examined and followed before applying nanostructures in hazard removal. The registration, evaluation, authorisation and restriction of chemicals (REACH) regulation provides a framework for managing chemical risks including nanostructures [
165]. Nanostructures used for hazard removal in aquaculture must be registered and authorized accordingly. Risk assessment protocols should evaluate potential impacts on aquatic organisms, including toxicity and bioaccumulation. International Organization for Standardization (ISO) 12878:2012 provides environmental monitoring guidelines in aquaculture, applicable to monitoring nanostructures’ impact [
166]. ISO Technical Specifications (ISO/TS) 22002-3:2011 establishes food safety prerequisites to ensure that nanostructures do not compromise aquatic product safety [
167]. Although nanostructure-based detection does not require the addition of nanostructures into foods, proper disposal is encouraged. ISO 45001:2018 requires safety management for workers handling nanostructures [
168]. Regular updates on regulations and risk assessments are crucial for protecting aquatic ecosystems and public health.
PDF
(6621KB)
