Applications of nanomaterials with enzyme-like activity for the detection of phytochemicals and hazardous substances in plant samples

Abstract

Plants such as herbs, vegetables, fruits, and cereals are closely related to human life. Developing effective testing methods to ensure their safety and quantify their active components are of significant importance. Recently, nanomaterials with enzyme-like activity (known as nanozymes) have been widely developed in various assays, including colorimetric, fluorescence, chemiluminescence, and electrochemical analysis. This review presents the latest advances in analyzing phytochemicals and hazardous substances in plant samples based on nanozymes, including some active ingredients, organophosphorus pesticides, heavy metal ions, and mycotoxins. Additionally, the current shortcomings and challenges of the actual sample analysis were discussed.

Background

Plants, such as Chinese herbal medicines (CHMs) and edible plants, play crucial roles in human life [1]. The bioactive components in plants, namely phytochemicals, are important for human health and disease prevention. They usually have antioxidant capacity that protects cells from oxidative stress, as well as anti-inflammatory, antibacterial, and anti-tumor activities to prevent diseases like cancer, inflammatory bowel disease, and metabolic syndrome [2,3,4]. To date, various analytical techniques have already been used to analyze phytochemicals, including high-performance liquid chromatography (HPLC), mass spectrometry (MS), capillary electrophoresis (CE), gas chromatography (GC) [5, 6], etc. Although these methods are accurate and sensitive, sample handling is sophisticated with long analysis time and high cost. Thus, the development of novel, rapid, and selective approaches for the analysis of phytochemicals in plants and their extracts is of great importance for quick and on-site detection.

Furthermore, plants are frequently exposed to a variety of chemicals that are harmful to the human body, impacting their usability. For instance, inappropriate discharge of industrial wastewater leads to the accumulation of heavy metal ions in the soil, which will inevitably be absorbed by plants during planting. Excessive intake of heavy metal ions is prone to cause neurological disorders, kidney and liver damage, cardiovascular disease, and cancer [7, 8]. Therefore, the Chinese Pharmacopoeia (2020 edition) has set limits for heavy metals in Chinese medicinal materials and tablets of plant species: arsenic (2 mg/kg), cadmium (1 mg/kg), copper (20 mg/kg) lead (5 mg/kg), and mercury (0.2 mg/kg) [9]. In addition, organophosphorus pesticides (OPs) are widely used to protect plants from pests during cultivation, giving rise to the presence of excessive pesticide residues [10]. Once entering the human body, they irreversibly inhibit cholinesterase activity, posing a hazard to the cardiovascular, nervous, and respiratory systems [11]. In response, the Ministry of Agriculture and Rural Affairs of China has established the maximum residue limits (MRLs) of OPs in food. For example, apples’ MRLs of dichlorvos, glyphosate, and chlorpyrifos are 0.1, 0.5, and 1 mg/kg, respectively (GB 2763–2021) [12]. Furthermore, it should also be noted that plants may become contaminated by certain toxins during storage, such as mycotoxins. These toxins are highly carcinogenic and difficult to be completely removed during processing because of their thermal stability [13]. The current analytical methods for detecting heavy metal ions, OPs, and mycotoxins, including HPLC, GC–MS, atomic absorption spectrometry (AAS), inductively coupled plasma mass spectrometry (ICP-MS), and ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS), require complex pretreatment and specialized operations [8, 10, 14]. Therefore, there is a necessity to develop simple, efficient, and sensitive methods to detect these hazardous substances in plants for on-site and rapid inspection.

Nanozymes are a type of nanomaterials that exhibit enzyme-like activities, including metals (Au, Ag, Pt, Pd, etc.), metal oxides (CeO₂, Fe₃O₄, Mn₃O₄, CuO, etc.), carbon-based compounds (carbon nanotubes, graphitic carbon nitride, carbon dots, etc.), and other nanomaterials (e.g., metal–organic frameworks (MOFs), covalent organic frameworks (COFs), metal sulfides, etc.) [15,16,17]. The enzyme-like activity of currently reported nanozymes can be mainly divided into two categories, oxidoreductase such as peroxidase (POD), oxidase (OXD), laccase (LAC), and superoxide dismutase (SOD), and hydrolase such as nuclease, esterase, phosphatase, and protease. Table 1 summarizes the functions of commonly reported nanozymes. Nanozymes have been widely used in the field of biomedicine, environmental monitoring, and food safety for their remarkable advantages of high stability, low cost, controllable activity, and easy storage [18, 19]. Most significantly, the nanozyme-based sensor offers the advantage of a shorter detection time that can fulfill the requirements of real-time detection [20,21,22]. For instance, Wang et al. [23] developed a manganese-based nanozyme that enabled rapid quantitative analysis of glutathione within 1 min. Xu et al. [24] synthesized copper-cobalt bimetallic nanozymes and combined with a smartphone and hydrogel kit to achieve real-time monitoring of perfluorooctane sulfonate (PFOS) in lake water. The approach offers a simpler instrument and quicker build-up compared to traditional methods like HPLC. Herein, this review aims to summarize recent advancements in applying nanomaterials with enzyme-like activity to detect phytochemicals and hazardous substances in plants (Fig. 1). Firstly, the application of nanozymes for detecting active phytochemicals was introduced, including gallic acid, tannic acid, ascorbic acid, rutin, atropine, quercetin, astragaloside-IV, and licorice. Secondly, advancements in the utilization of nanozymes for detecting hazardous substances in plants were presented, such as organophosphorus pesticides, heavy metal ions, and mycotoxins. Finally, the challenges and prospects in nanozyme-based detection of plant samples were discussed. This paper may provide useful information for readers to understand the design, performance, and application of nanozymes, to develop efficient, rapid, highly sensitive, and selective methods for detecting target components in actual plant samples.

Table 1 Summary of functions of currently reported nanozymes

Fig. 1

Review of nanozymes-based detection of phytochemicals and hazardous substances in plants

Detection of phytochemicals

Phytochemicals are biologically active secondary metabolites produced by plants for self-protection, including carotenoids, polyphenols, alkaloids, saponins, and others [25], which are obtainable from a variety of sources, such as herbs, vegetables, fruits, and teas [26]. Most of them exhibit potent antioxidant activity and contribute to reducing the risks of heart disease, cancer, diabetes, and other diseases [25, 27]. However, plants are intricate systems containing a multitude of substances, making it challenging to achieve specific analysis and identification of the target phytochemicals. Table 2 summarizes some of the studies on the detection of phytochemicals in plants by nanozyme-based methods.

Table 2 Summary of detection of phytochemicals in plants based on nanozymes

Direct detection

Gallic acid (GA) and tannic acid (TA) are a class of natural phenolic compounds widely found in fruits and teas with various biological activities, such as antioxidant, anticancer, anti-mutagenesis, and antiviral [28,29,30,31]. The nanomaterials with POD-like and OXD-like activities can catalyze the oxidation of the substrate 3, 3’, 5, 5’-tetramethylbenzidine (TMB) to generate blue oxidized TMB (ox-TMB) in the presence of H₂O₂ and O₂, respectively. GA and TA can inhibit the oxidation of TMB due to their antioxidant property, realizing colorimetric detection of them. Besides, instead of complicated pretreatment, these active ingredients can be directly detected in the real samples through a simple water extraction. Perovskite is a type of transition metal oxide, some of which possess splendid catalytic activity. Chen et al. [29] developed a simple colorimetric method to detect GA based on the POD-like activity of LaFeO₃ microspheres, which is a typical perovskite, with a linear range of 0.67–40.8 µM and a limit of detection (LOD) of 0.4 µM. In addition, the established method was used in the determination of GA in diet tea, green tea, and pharyngitis tablets with good recoveries of 95.65–102.10% and RSD (n = 3) less than 4.00%. The activity of nanozymes plays a vital role in detecting phytochemicals, which can affect the detection sensitivity. Combining carbon-based materials with perovskite can enhance their catalytic performance. Liu et al. [32] synthesized the heterojunctions composed of strontium titanate (SrTiO₃) and reduced graphene oxide (rGO), which facilitate photo-generated charge transfer under ultraviolet irradiation, resulting in an excellent POD-like activity. It is noted that the affinity for TMB of SrTiO₃-rGO composites is 19 times higher than that of natural horseradish peroxidase (HRP). Meanwhile, the colorimetric quantitative detection of TA shows a lower LOD of 0.056 µM, which has been successfully applied to detect TA in green tea and Oolong tea.

Atomic doping is also one of the methods to enhance enzyme-like activity. Furthermore, oxygen vacancies (OVs) are a kind of metal oxide defects, which are formed by the detachment of oxygen from the lattice of metal oxides in a specific external environment (e.g., high temperature). OVs can provide rich active sites and high surface energy to improve the catalytic activity of nanozymes [33]. Zhou et al. [28] prepared a raspberry-like nitrogen-doped Mn₃O₄ nanospheres (N-Mn₃O₄ NSps) with OVs, which exhibited enhanced OXD-like activity (Fig. 2). The senor based on N-Mn₃O₄ NSps showed excellent reproducibility, stability, and interference resistance for detecting GA with a linear range of 5–30 µM and a lower LOD of 0.028 µM, which is feasible for the detection of GA both in green tea and black tea with the RSD (n = 5) within 3.27%. Furthermore, a platform based on the smartphone was implemented for GA detection with a LOD of 0.047 µM.

Fig. 2

Schematic diagram of detecting GA based on N-Mn₃O₄ NSs. Reprinted with permission from [28]

In addition, the electrochemical assay has also been exploited for the quantification of GA. Based on gold nanoparticles, Chen et al. [34] developed a peptide-modified dual mimetic enzyme sensor for the detection of GA. The construction mechanism relied on the active center of the methanobactin (Mb) structure that can capture Cu(II), resulting in the coordinated complex Mb(Cu^II) with polyphenol oxidase (PPO)- and POD-like activities. After the addition of GA, the sensor with surface-modified gold nanoparticles and Mb(Cu^II) exhibited a high oxidation peak with a peak potential of 0.79 ± 0.05 V. Subsequently, the developed method was employed to detect GA in three real samples, including grapes, oranges, and black tea, with recoveries of 96.76–100.95% and RSD (n = 3) less than 5%.

To explore additional response mechanisms is an effective approach to improve the detection selectivity. The 2,3,6,7,10,11-Hexahydroxytriphenylene (HHTP) is a highly conjugated triol ester that can coordinate with a metal-based node to form two-dimensional porous expansion frameworks known as metal catecholates (M-CATs). Inspired by the structure of M-CATs, Wu et al. [31] prepared a Fe-HHTP amorphous nanomaterial with POD-like activity through an one-step self-assembly strategy. The colorimetric method based on Fe-HHTP can rapidly detect TA within the linear range of 0.5–100 µM with a LOD of 0.5 µM, which was successfully used to measure TA content in tea and red wine samples. Remarkably, the inhibition of TA on the color reaction was resulted not only from its antioxidant ability but also from the formation of a Fe³⁺-TA complex. However, GA and AA still exhibited certain interference on the detection of TA. To further address this issue, Wu et al. [35] developed a colorimetric/photothermal dual-mode analysis method for TA detection based on the light-enhanced POD-like activity and high photothermal property of CuS hollow nanocages (CuS HNCs) (Fig. 3). TA inhibited the oxidation of TMB and effectively captured the thermal holes generated by CuS HNCs under NIR irradiation, which suppressed the reaction system’s photothermal effect. The established method exhibited better selectivity and higher interference resistance from GA and AA.

Fig. 3

Schematic mechanism for dual-mode detection of TA based on the CuS HNC probe. Reprinted with permission from [35]

However, most of the above methods failed to identify GA or TA with absolute specificity because of the interference of other antioxidants in the samples. Alternatively, similar methods were applied to detect total antioxidant capacity (TAC) in the actual samples. He et al. [36] designed a Pd–Pt-Ru nanozyme with good POD-like activity, which was used to detect ascorbic acid (AA) and H₂O₂ in the ranges of 2–12 µM and 5–40 mM with the LOD values of 1.13 µM and 2.79 mM, respectively. The method was applied in the evaluation of TAC of drinks (iced tea and green tea), foods (orange, lemon, and tomato), and herbs (Cornus officinalis, Cynanchum otophyllum, Dioscorea bulbifera, and Eriobotryae Folium). The results demonstrate that orange and C. officinalis have a higher TAC. Based on the OXD-like activity of cobalt oxyhydroxide (CoOOH) nanorings, Zhang et al. [37] developed a colorimetric sensor with smartphone assistance for the detection of antioxidants in green tea (Fig. 4). The detection mechanism is the decomposition of CoOOH nanorings into Co²⁺ after the addition of antioxidants, resulting in a decrease of catalytic activity. The established method exhibited high sensitivity with a LOD of 0.025 µM. Moreover, a smartphone can be used as a readout, and the content of total antioxidants in green tea was measured to be 2.55 µM, which is close to the result of Folin’s method. Meanwhile, Murilo et al. [38] synthesized the manganese dioxide/graphene quantum dot (MnO₂/GQD) composites with excellent OXD-like activity, which was applied to detect the total antioxidants in fresh lemon juice, black tea, and mango juice, with recovery values of 95–105%. It is noteworthy that the system can differentiate different antioxidants by treating the obtained data through principal components analysis (PCA).

Fig. 4

Schematic diagram of detecting antioxidants based on CoOOH nanoring. Reprinted with permission from [37]

Detection after sample pretreatment

Some active ingredients, such as quercetin and rutin, are insoluble in water, so long-time alcohol extractions are needed. During the preparation of real samples, Davoodi-Rad et al. [39] dried the vegetable samples at 60 °C for 4 h, then ground them into powder. A portion of the powder was mixed with methanol and stirred for 24 h, finally followed by filtration, washing, and dilution. Additionally, several plants, like Datura stramonium, contain diverse components. Therefore, the detection of specific components in them requires complex extraction processes. The preparation of Datura samples required drying and degreasing, which was first extracted with methanol and filtered, and then rotary evaporated to remove the solvent. Following ultrasonication, the samples were further extracted twice with dichloromethane (DCM). This process involved several separation steps, pH adjustments, and drying steps [40].

Quercetin, which is a type of naturally polyphenolic flavonoid compound, is one of the active ingredients in many frequently used CHMs and natural products, such as Ginkgo biloba, licorice, and onions. Quercetin has various properties, including antioxidant, anti-cancer, hypoglycemic, and liver-protective [41]. Cao et al. [42] synthesized the CeO₂/Co₃O₄@N-doped hollow carbon microspheres (CeO₂/Co₃O₄@NCH) through a self-template method, which exhibited excellent POD-like activity due to its larger surface area, pore-like structure, and OVs. Based on the reduction property of quercetin, a facile, fast, and cheap sensor was established to detect it with a linear range of 7–22 µM and a LOD of 1.19 µM. In addition, the sensor was applied to analyze quercetin in Yinxingye Dispersible Tablets, showing satisfactory recoveries. Moreover, some nanozymes based on LAC-like activity have also been developed for the quantitative analysis of quercetin. LAC is a copper-containing polyphenol oxidase that catalyzes polyphenols and polyamines to produce colored ortho-quinone [39, 43]. Davoodi-Rad et al. [39] synthesized the Cu-TA nanosheets (Cu-TA NSs) with LAC-like activity to detect quercetin in treated vegetable samples. Firstly, Cu-TA NSs can oxidize quercetin to generate ortho-quinone. Additionally, the addition of the surfactant cetyltrimethylammonium bromide (CTAB) reacted with quercetin by supramolecular interaction, further promoting the oxidation of quercetin. The developed method showed good selectivity to detect quercetin with a lower LOD of 0.064 µM. Then, it was used to detect the quercetin content in red onion, green pepper, and drill samples, and the results are in consistent with that of HPLC analysis.

Nanozyme-based detections for total polyphenols have also been developed. For instance, Rashtbari et al. [44] synthesized molybdenum trioxide nanoparticles through Argon cold plasma surface modification (Ar-MoO₃NPs), which exhibited enhanced POD-like activity. The prepared Ar-MoO₃NPs can oxidize non-fluorescent terephthalic acid into high-fluorescence emission compounds in the presence of H₂O₂. In contrast, polyphenols can cause aggregation of Ar-MoO₃NPs and act as free radical scavengers, leading to the quenching of fluorescence. Therefore, a fluorescence method was developed to detect polyphenols with high specificity, which was successfully used to detect total polyphenolics in apple, orange, and grape samples.

Rutin, which has antioxidant, anticancer, vasoprotective, and neuroprotective properties [45], is a polyphenolic flavonoid compound and can be hydrolyzed to produce quercetin. Davoodi-Rad et al. [46] developed a colorimetric method for detecting rutin based on the LAC-like activity of Cu-guanosine nanorods (Cu-Guo NRs), which can oxidize rutin, resulting in a color change from light green to dark yellow. The established strategy has a broad linear range of 0.77–54.46 µM and a LOD of 0.114 µM. Then, it was successfully used to detect rutin in propolis dry extract and rutin-containing dietary supplement tablets, with contents of 9.42% and 18.38 mg per tablet, respectively. Covalent triazine framework (CTF) is a special class of COFs with a triazine ring in its structure. Based on the advantage of chemiluminescent (CL) detection with high sensitivity, Tan et al. [47] prepared a CTF-1 with OXD-like activity, which can oxidize luminol to produce intense CL in the presence of O₂ (Fig. 5). Whereas the intensity of CL decreased with the increase of rutin concentration, therefore, a very sensitive CL method can be established for detecting rutin with a LOD of 0.015 µM. Compared with the results of HPLC, the developed CL method is reliable for detecting rutin both in tablets and treated Flos Sophorae Immaturus samples.

Fig. 5

Schematic diagram of CL of luminol based on CTF-1. Reprinted with permission from [47]

Atropine is an alkaloid that can be used to dilate the pupil, alleviate spasms, and serve as an antidote to organophosphorus pesticides. Datura is a poisonous plant but contains abundant active chemicals, including phenolics, steroids, acyl sugars, amides, and alkaloids. Therefore, tedious sample pretreatment is necessary to detect atropine content in Datura plants. Mahmoudi et al. [40] synthesized a series of Fe₃O₄ and bimetal-organic framework Zn/Mg (Fe₃O₄@MOFs) composites for the detection of atropine extracted from two Datura samples through liquid–liquid extraction. The experimental results show that the Fe₃O₄@MOF/Dextrin composite exhibited the highest POD-like activity, which was primarily attributed to the cooperative interaction of dispersed Fe ions between Zn and Mg metals in the MOF and dextrin layers. In the presence of the material and H₂O₂, terephthalic acid was oxidized to 2-hydroxy terephthalic acid, emitting fluorescence at 425 nm. This oxidation process can be inhibited by atropine, allowing the fluorescence detection of atropine with the LOD value of 2.27 μg/L. To further improve the sensitivity of atropine, the group [48] developed a CL method to detect atropine based on the Fe₃O₄@MOF composite, which can oxidize luminol, producing high-intensity CL (Fig. 6). While atropine can bind with the Fe₃O₄@MOF composite, leading to a significant reduction of CL intensity. Compared with the previous method, the sensitivity was considerably improved and the LOD was as low as 0.02 μg/L.

Fig. 6

Schematic diagram of detecting atropine based on Fe₃O₄@MOFs. Reprinted with permission from [48]

Specific recognition strategy

Molecularly imprinted polymers (MIPs) are formed by polymerizing monomers in the presence of a template molecule. After removing the template molecule, MIPs can be used to bind template molecule specifically, similar to the interaction between an antibody and an antigen [49, 50]. Therefore, MIP can extract template molecules from complicated samples and shield interference from other substances, improving the assay selectivity [51]. As one of the main active ingredients in Huangqi (Astragalus membranaceus), Astragaloside-IV (AS-IV) has many pharmacological activities, including enhancing immunity, antivirus, anti-stress, antifibrosis, and protecting the heart [52, 53]. The AS-IV is usually detected by HPLC in combination with other methods due to its weak ultraviolet absorption, such as pulsed amperometric detection and evaporative light scattering detection (ELSD) [53, 54]. Chen et al. [51] innovatively combined MIP with CuO nanoparticles (CuO NPs) and polydopamine (PDA) to synthesize MIP@PDA/CuO NPs with POD-like activity for the detection of AS-IV. AS-IV can specifically bind to the surface imprinted cavity to prevent the entry of H₂O₂ and TMB, inhibiting the catalytic process (Fig. 7). Eventually, the established new colorimetric method for the detection of AS-IV showed high selectivity and the linear range was 0.000341–1.024 mg/mL with a LOD of 0.000991 mg/mL. Additionally, it was applied to detect the content of AS-IV in Huangqi Granules and Ganweikang Tablets, and the results are similar to that measured by HPLC-ELSD.

Fig. 7

Schematic diagram of preparing MIP@PDA/CuO NPs (A) and detecting AS-IV (B). Reprinted with permission from [51]

Licorice (Glycyrrhiza uralensis) is a CHM with diverse functions, such as anti-inflammatory and detoxification. Among licorice active ingredients, liquiritin and glycyrrhizic acid are the indicators for authenticating licorice, while licochalcone A and isolicoflavonol are the indicators for evaluating its quality. Thus, the simultaneous detection of these four active substances is of great significance. The sensor array consists of a series of cross-response sensing units rather than a specific receptor, which can detect and discriminate structurally similar components or complex mixtures through pattern recognition [55]. Based on three iron oxide nanozymes (Fe₂O₃, Fe₃O₄, and histidine (His)-Fe₃O₄) with POD-like activity, Yuan et al. [56] constructed a colorimetric sensor array for the detection of four licorice active substances (Fig. 8). Different active ingredients inhibited the catalytic activity of different iron oxides to various degrees, and the developed colorimetric sensor successfully identified and distinguished the four licorice active substances in real licorice samples in the concentration range of 1–200 µM.

Fig. 8

Schematic diagram of detecting four licorice active substances based on the colorimetric sensor array. Reprinted with permission from [56]

Although nanozyme-based detection methods have the advantages of simplicity, rapidity, and high sensitivity, there were limited number of methods have been developed for the detection of phytochemicals in natural products using nanozymes. Furthermore, due to the complexity of real samples, there are still difficulties in achieving specific detection, which requires sample pretreatment or combining with special methods such as MIP. Therefore, the design and preparation of nanozymes with high selectivity to solve the problem of complex sample matrix in phytochemical analysis remain in the exploratory stage.

Detection of hazardous substances

Detection of heavy metal ions

Due to their bioaccumulation properties, heavy metal ions can reach very high levels through the diet, thereby harming human health [8]. Common heavy metal elements include arsenic (As), lead (Pb), mercury (Hg), copper (Cu), chromium (Cr), cadmium (Cd), iron (Fe), etc. [57]. As shown in Table 3, there are several studies on the detection of As and Pb in plants using nanozymes.

Table 3 Summary of detection of heavy metal ions in plants based on nanozymes

Arsenic ion

Compared with organic arsenic, inorganic arsenic exhibits higher toxicity. Excessive intake of it can cause skin and respiratory diseases, nerve poisoning, organ failure, and even cancer, which may be resulted from its interaction with enzymes in the human body and excess generation of reactive oxygen species (ROS) [58,59,60,61]. Inorganic arsenic includes arsenic trivalent (As(III)), arsenic pentavalent (As(V)), and elemental arsenic, while As(III) is more toxic than As(V) as it can bind to sulfhydryl groups with higher affinity, inhibiting the activity of various proteins [59, 61]. Wang et al. [59] developed a colorimetric method to detect As(III) (Fig. 9). They synthesized different shapes of Mn₃O₄ NPs with OXD-like activity, while the octahedral one possessed the strongest As(III) adsorption capacity. Furthermore, arsenic adsorption made Mn⁴⁺/Mn³⁺ reduced to Mn²⁺, which can catalyze O₂ to produce oxygen radicals, further oxidizing TMB with the solution turning to yellow color. Based on As(III)-adsorption enhancing the catalytic activity of octahedral Mn₃O₄ NPs, a colorimetric method was established and achieved the detection of As(III) in the wheat sample with a LOD of 1.32 μg/L.

Fig. 9

Schematic diagram of detecting As (III) based on octahedral Mn₃O₄ NPs. Reprinted with permission from [59]

As(V) can inhibit the catalytic activity of acid phosphatase (ACP), which can catalyze the hydrolysis of ascorbic acid 2-phosphate (AAP) to produce AA. Therefore, several studies have utilized nanozymes and ACP to design enzyme-cascade reactions for As(V) detection. Xu et al. [58] prepared an ACP and hemin-loaded multifunctional Zn-based metal–organic framework (ACP/hemin@Zn-MOF) for the detection of As(V). Hemin exhibited POD-like activity, which can catalyze the oxidation of o-phenylenediamine (OPD) to form a fluorescent product (564 nm) and weaken its intrinsic fluorescence (452 nm) owing to the inner filter effect. After the addition of AAP, the generated AA will competitively suppress the oxidation of OPD, causing a decrease in the fluorescence intensity at 564 nm and a recovered fluorescence at 452 nm. The inhibitory effect of As(V) on ACP enabled the fluorescence signal to be reversed again, realizing a ratio fluorescence detection of As(V) with a linear range of 3.33–300.00 μg/L and a LOD of 1.05 μg/L. Moreover, the method was successfully applied in the analysis of As(V) and total arsenic in rice samples, with recovery rates ranging from 95 to 105%. Similarly, Wang et al. [60] developed a colorimetric method to detect As(V) utilizing the OXD-like activity of Ce(IV) coordination polymer nanoparticles. With the addition of ACP and AAP, the produced AA can not only restrain the oxidation of TMB but also reduce Ce⁴⁺ to Ce²⁺, inhibiting the enzyme-like activity of the material. Therefore, As(V) can be detected by inhibiting ACP and restoring the TMB color reaction. The method displays a high sensitivity and was used to analyze As(V) in rice samples.

Lead ion

Pb is the second most toxic heavy metal after As with bioaccumulation and persistence [62,63,64,65,66]. Low doses of Pb²⁺ have an impact on the physical and mental health of infants and young children, causing developmental disorders, brain damage, psychiatric disorders, etc. [67]. Tang et al. [66] synthesized the layered WS₂ nanosheets with POD-like activity through a simple ultrasonic stripping method, which was employed to detect Pb²⁺ in wheat samples. Pb²⁺ blocked the electron transfer between WS₂ and H₂O₂, and then prevented the oxidation of TMB, resulting in a significant decrease of absorbance at 650 nm. For Pb²⁺ detection, the linear range of the method was 0.015–0.24 nM with a low LOD of 0.012 nM. Using Pb²⁺-dependent receptors (e.g. DNAzyme) is a good option to achieve a high selective detection. Si et al. [65] developed an electrochemical method to monitor Pb²⁺ in vegetable samples based on a porphyrin-functionalized metal–organic framework (porph@MOF) and Pb²⁺-dependent DNAzyme. As shown in Fig. 10, DNA2 was immobilized on an AuNPs-modified glassy carbon electrode via the Au–S bond. It can be specifically cleaved by Pb²⁺ to generate a short DNA2 fragment, which was further hybridized with porph@MOF-DNA1 through base pairing. Subsequently, the porph@MOF with POD-like activity oxidated OPD in the presence of H₂O₂, producing the electrochemical signal. The established method exhibited excellent selectivity and high sensitivity with a LOD of 5 pM. However, although cleavage and hybridization of DNA2 can be finished in one step, the long incubation time (80 min) did not fulfill the requirement of rapid detection. Colorimetric and surface-enhanced Raman spectroscopy (SERS) are commonly regarded as rapid analytical methods [68]. Gold nanoparticles (AuNPs) are widely studied for their enzyme-like and SERS properties [69, 70]. Furthermore, several studies have substantiated that carbon dots (CDs) facilitate improving SERS signals and catalytic activity of AuNPs [71,72,73]. For example, Cui et al. [64] prepared the Fe-doped norepinephrine-based CDs through a one-step microwave digestion method, which were self-assembled with Prussian blue (PB) to obtain Fe, NA-CDs/PB with POD-like activity. They also utilized the reducing property of CDs to synthesize the AuNPs. Based on the inhibition of Pb²⁺ on the POD-like activity of Fe, NA-CDs/PB and AuNPs, the SERS and colorimetric dual-mode sensor was constructed with the LODs of 0.024 nM and 0.015 nM, respectively. Finally, the sensor was successfully applied to detect Pb²⁺ in Salvia miltiorrhiza, Astragalus membranaceus, Barley Yellow, and pomegranate peel with good recovery of 90.4–108.9% and RSD of 2.6–4.7%.

Fig. 10

Schematic diagram of detecting Pb.²⁺ based on DNAzyme and porph@MOF. Reprinted with permission from [65]

Other ions

Ingestion of inorganic mercury may cause neurological symptoms (including mental retardation, vision and hearing loss, language disorders, and memory loss), as well as cognitive and motion disorders, etc. [74, 75]. Recently, single-atom nanozymes (SAzymes) with ultra-high atomic utilization, excellent stability, and remarkable catalytic activity have been continuously studied [75,76,77]. Ge et al. [75] developed a novel colorimetric strategy for Hg²⁺ assay using cysteine and single-atom Cu-C-N nanozymes (SACu-C-N). The prepared SACu-C-N exhibited OXD-like activity, catalyzing the oxidation of TMB to blue ox-TMB. However, the ox-TMB was reduced after the introduction of cysteine. Since Hg²⁺ possesses a strong affinity for thiol groups of cysteine, it can turn the solution to blue color again. Therefore, a simple, sensitive, and selective colorimetric method was established and applied in the analysis of Hg²⁺ in cabbage samples.

Cu is an essential trace element and is influential in the metabolic process as a cofactor or structural component of many natural enzymes [78,79,80]. However, excess Cu²⁺ suppresses the activity of some essential enzymes, causing serious side effects, such as neurodegenerative disease, liver damage, and even cancer [79, 81]. In addition, high levels of Cu also damage photosynthesis and then inhibit plant growth [78]. Xiong et al. [80] synthesized a CTF through a simple and rapid microwave-enhanced high-temperature ionothermal method. Interestingly, CTF possessed weak POD-like activity, but Cu²⁺ can act as the active catalytic center and a bridge for electronic transfers between the substrate and CTF, resulting in enhanced catalytic activity. The LOD value of the developed method was 1.25 nM, and was applied in the quantification of Cu²⁺ in Chinese water chestnuts and eggplants, with recoveries of 96.0–105.0%.

Most of the sensors can detect only a single heavy metal, and the simultaneous detection of multiple heavy metal ions is still a challenge. Song et al. [76] designed a time-resolved sensor to detect Cr⁶⁺ and Fe³⁺ based on their difference in enhancing the single-atom Ce–N–C nanozyme’s OXD-like activity. In the presence of Fe³⁺ and Cr⁶⁺ alone, the solution turned blue after 30 s and 60 s, respectively, while the former faded after 5 min. The solution turned blue in 30 s and did not fade when both of them were presented. Therefore, the constructed sensor was feasible for the simultaneous detection of Fe³⁺ and Cr⁶⁺ in actual samples. By utilizing gold nanoclusters (AuNCs) as sensing elements, Li et al. [82] developed a colorimetric sensor array for identifying five heavy metal ions (Hg²⁺, Pb²⁺, Cu²⁺, Cd²⁺, and Co²⁺) at a concentration down to 0.5 μM, which was successfully used to recognize multiple heavy metal ions in Lonicera japonica and Chrysanthemum morifolium samples.

Detection of mycotoxins

Mycotoxins are secondary metabolites generated by filamentous fungi and are extensively found in maize, wheat, rice, peanuts, and other cereals, which include aflatoxins, ochratoxins, fumonisins, zearalenone [14, 83], etc. Even at low concentrations, they are nephrotoxic, immunotoxic, teratogenic, mutagenic, and carcinogenic [84, 85]. According to the Chinese Pharmacopoeia, the total aflatoxin content in Chinese medicines should not exceed 10 µg/kg, and zearalenone should not exceed 500 µg/kg [86]. At present, a number of nanozyme-based immunoassays have been reported for detecting mycotoxins (Table 4).

Table 4 Summary of detection of mycotoxins in plants based on nanozymes

Aflatoxin b1

Aflatoxins include aflatoxin B1, B2, G1, and G2 [87]. Among them, aflatoxin B1 (AFB1) is the most toxic one with potent hepatocarcinogens, which was classified as a Group 1 carcinogen as early as 2002 for it can induce formatting DNA adducts, leading to hepatoma [88,89,90]. Apart from this, AFB1 is also associated with malnutrition, growth impairment, and immune inhibition [91,92,93]. Lateral flow immunoassay (LFIA) is a real-time analysis method on paper-based equipment. The principle is mainly based on the competitive binding of the target analyte and the fixed antigen on the detection line to the antibody [94]. Owing to its simplicity, rapidity, and low cost, LFIA has become an attractive immunoassay for AFB1 analysis. However, the limited sensitivity hinders LFIA's practical applications [95,96,97]. As mentioned, nanozymes can catalyze the formation of chromogenic substrates for signal amplification to improve the sensitivity of a detection method. Cai et al. [95] prepared MnO₂ nanosheets with excellent OXD-like activity as signal labels conjugated with antibodies to detect AFB1. Using MnO₂ catalyzing TMB to produce clear color signals, the method achieved sensitive detection of AFB1 with a LOD of 15 pg/mL and a wide linear range of 0.01–150 ng/mL. Compared to antibodies, aptamers are more stable and flexible in labeling. Therefore, aptamer-mediated LFIA is a promising approach to realize a highly sensitive detection. Zhu et al. [97] designed a PDA-modified nanozyme (CuCo@PDA) with abundant amide groups that can be coupled to AFB1 aptamers via a condensation reaction. Based on the POD-like activity of CuCo@PDA, a reliable and ultrasensitive method combined with a smartphone was established for AFB1 detection with a LOD of 2.2 pg/mL. Moreover, it was successfully applied to detect AFB1 in the peanut, corn, and wheat samples with different contamination levels.

The enzyme-linked immunosorbent assay (ELISA) is a widely used immunoassay. In a typical ELISA assay, antigens (analytes) first bind to antibodies immobilized on a well plate, then forming an antibody-antigen–antibody sandwich with enzyme-labeled antibodies (commonly HRP). After washing steps, HRP catalyzes the added substrate, resulting in a color change [98]. Likewise, the combination of nanozymes can effectively improve the relatively low sensitivity of ELISA [77, 99]. Guo et al. [77] developed a nanozyme-linked immunosorbent assay (NLISA) by utilizing Fe–N–C SAzymes to replace HRP for quantitative detection of AFB1 in peanut samples. Unlike the traditional antibody-antigen–antibody “sandwich” type of detection mechanism, the method immobilized antigens on the well plate, achieving the rapid and sensitive detection of AFB1 with a LOD of 3.3 pg/mL. The coupling of nanozymes with bio-enzymes induces a dual signal amplification through an enzyme cascade to further increase the sensitivity. Lai et al. [99] found that copper hexacyanoferrate nanoparticles (CHNPs) with OXD-like activity can be rapidly produced by simply mixing potassium hexacyanoferrate(III) (K₃[Fe(CN)₆]) with Cu(II). However, AA produced by the hydrolysis of ALP on ascorbic acid 2-phosphate (AAP) can reduce Fe (III) to Fe (II) and then inhibit the formation of CHNPs. Therefore, employing AuNPs coupled to ALP as enzyme labels in ELISA and integrating with the production process of CHNPs, a highly sensitive colorimetric immunoassay for the determination of AFB1 was constructed with a LOD of 0.73 pg/mL. Finally, the developed method was used to detect AFB1 in spiked and naturally contaminated peanut samples.

The single signal mode is prone to false-negative/positive results caused by differences in operating conditions and environment. In contrast, multi-mode detection can offset interferences, reduce false results through self-correction, and yield more precise outcomes [90, 100]. Huang et al. [101] developed a colorimetric/photothermal dual-mode immunoassay method based on Pt supported on nitrogen-doped carbon (Pt-CN) for monitoring AFB1 in peanut samples. After competitive immunoreactivity of glucose oxidase (GOx)-labeled antigen with AFB1, the GOx loaded on the well plates can catalyze the formation of H₂O₂ from glucose. Subsequently, the Pt-CN with POD-like activity can oxidize TMB to blue ox-TMB, producing a colorimetric signal. On the other hand, the ox-TMB, as a photothermal agent, can convert light to heat under near-infrared (NIR) irradiation, generating a photothermal signal. The LOD values are 0.22 pg/mL and 0.76 pg/mL for the colorimetric and photothermal assays, respectively. The fluorescence method is regarded as one of the most sensitive of the optical methods. Lu et al. [90] designed a photothermal/colorimetric/fluorescent multimodal NLISA to portably and ultra-sensitively detect AFB1. As shown in Fig. 11, magnetic nanoparticles (MNPs) combined with aptamers were immobilized on the well plates via antibody and AFB1. In the presence of K₄[Fe(CN)₆] and HCl, MNPs as precursors can form Prussian blue nanoparticles (PBNPs) that possessed both excellent POD-like activity and photothermal effect. Particularly, MNPs also acted as the quencher to decrease the fluorescence of the dye (Cy5), which was restored upon the formation of PBNPs, enabling fluorescence detection of FAB1, with an extremely low LOD of 0.54 fg/mL. The established strategy was feasible for the qualitative and quantitative determination of AFB1 in the actual samples on the spot.

Fig. 11

Schematic diagram of multimode detecting AFB1. Reprinted with permission from [90]

Ochratoxin A

Ochratoxins are a series of mycotoxins generated by Penicillium and Aspergillus. Among them, ochratoxin A (OTA) is the most poisonous to human health. It possesses various toxicity in animals and humans, including nephrotoxicity, hepatotoxicity, immunotoxicity, teratogenicity, and carcinogenicity [100, 102, 103]. Recently, studies have shown that OTA is also a potent neurotoxin that is considered to be a causative agent of neurodegenerative diseases, and the brain is one of the main target organs of its damage [104]. As mentioned, dual-mode analysis has the advantage of better sensitivity and more accurate results, which was employed in most of the nanozymes-based detection methods of OTA [100, 105,106,107,108,109]. Li et al. [109] developed a SERS/colorimetric dual-mode method for the detection of OTA using Pd–Pt bimetallic nanocrystals (Pd–Pt NRs) conjugated with aptamers as recognition probes. Since the Pd–Pt NRs with POD-like activity can catalyze the oxidation of TMB to ox-TMB that exhibited a strong SERS signal, the SERS and colorimetric detection of OTA can be achieved with LODs of 0.042 nM and 0.097 nM, respectively. The developed method was applied to detect OTA in grape samples, and the results are in consistent with that of UPLC-MS/MS analysis. Zheng et al. [100] established a colorimetric/fluorescence immunoassay method for detecting OTA based on the OXD-like activity of cerium-based nanoparticles (CPNs(IV)) and the fluorescence properties of CPNs(III). The LODs of colorimetric and fluorescence methods are 0.962 ng/mL and 0.404 ng/mL, with recoveries in corn samples ranging from 99.12–102.60% to 97.60–103.55%, respectively. In addition, to improve the accuracy of visual judgments, colorimetric immunoassays with multiple color changes have also been reported. Gold nanomaterials have garnered attention for the color of their solutions, which largely depends on their shape and size [105, 107, 110]. Zhu et al. [110] synthesized octahedral Cu₂O nanoparticles with POD-like activity, which can oxidize TMB to TMB²⁺ in the presence of H₂O₂ and HCl. Based on the significant color change resulted from the etching of TMB²⁺ on gold nano bipyramids (Au NBPs), a multi-colorimetric immunoassay was developed to monitoring of OTA in millet samples with a LOD of 0.47 ng/L (Fig. 12).

Fig. 12

Schematic diagram of multi-colorimetric detecting OTA. Reprinted with permission from [110]

Zearalenone

Zearalenone (ZEN) is a kind of mycotoxin with estrogenic activity, which can compete with the natural estrogen, resulting in the reproductive dysfunction of animals [83]. In addition, ZEN may cause other toxic effects involving hepatotoxicity, immunotoxicity, genotoxicity, and carcinogenicity [83, 111,112,113,114111‒114]. Sun et al. [114] developed a colorimetric method for the detection of ZEN based on the inhibition of ZEN aptamer on the POD-like activity of AuNPs. After the addition of ZEN, the aptamer bound to it preferentially with the restoration of AuNPs activity. The LOD value of the method is 10 ng/mL and the recovery in spiked corn is in the range of 92%‒102%. Bimetallic nanoparticles are superior to monometallic nanoparticles in terms of catalytic activity [108, 109]. Liu et al. [113] synthesized encapsulated AuPt nanoparticles hydrogel by ZEN aptamer and complementary DNA as crosslinkers. In the presence of ZEN, it will preferentially combine with the aptamer, destroying the hydrogel structure and then releasing the AuPt nanozymes to complete the catalytic reaction. Therefore, a highly sensitive colorimetric method for the determination of ZEN was established with a LOD of 0.6979 ng/mL, which was applied to detect ZEN in corn and wheat samples. However, metal nanoparticles alone are prone to aggregation, resulting in reduced catalytic sites and lower catalytic activity. Anchoring it to a carrier material is an effective solution to this problem [91]. For example, utilizing Ti₃C₂T_x nanosheet as a carrier material, Huang et al. [112] prepared a Ti₃C₂T_x/AuNPs nanocomposite with enhanced POD-like activity, which can be used as an immunoprobe to detect ZEN. Employing Ti₃C₂T_x/AuNP to catalyze the oxidation of TMB and the strong NIR-driven photothermal effect of ox-TMB, the immunoassay achieved ultrasensitive colorimetric and photothermal dual-mode detection of ZEN with LODs of 0.15 pg/mL and 0.48 pg/mL, respectively. Furthermore, the dual-mode strategy was employed for the analysis of ZEN in three contaminated cereal samples, and the results are in good agreement with UPLC-MS/MS analysis.

Detection of organophosphorus pesticide

Organophosphorus pesticide exposure primarily causes chronic or acute toxicity in humans, plants, and animals through inhibiting cholinesterase activity and leading to acetylcholine accumulation [115]. The hazards of OPs on human beings are generally dominated by acute toxicity, which is manifested by a series of neurotoxic symptoms like sweating, tremors, confusion, speech disorders, and in severe cases, respiratory paralysis and even death [116]. As summarized in Table 5, the strategies for detecting OPs based on nanozymes are categorized as follows: (1) direct influence of OPs on the activity of nanozymes [117,118,119,120,121]; (2) nanozymes with organophosphorus hydrolase (OPH)-like or phosphatase-like activity hydrolyze OPs into products that produce signals [122,123,124,125,126,127,128,129,130,131]; (3) nanozymes combine with enzymes that can be inhibited by OPs such as acetylcholinesterase (AChE), alkaline phosphatase (ALP), and ACP [132,133,134,135,136,137,138,139,140,141]; (4) nanozymes combine with antibodies or aptamers [142, 143]. Since a review of comprehensive and systematic description of detecting OPs has been reported [10], this paper will not delve into too much detail.

Table 5 Summary of detection of organophosphorus pesticide in plants based on nanozymes

Challenges and prospects

Nanozymes can overcome some drawbacks of natural enzymes and exhibit higher catalytic activity. At present, some nanozymes have been designed for the analysis of plant samples, but there are still some limitations. To promote the development of nanozymes and their application in the detection of plant samples, the following challenges and prospects are proposed.

• (1) The nanozymes currently used for phytochemical detection are mainly nanomaterials with POD-like or OXD-like activity. The similar detection mechanisms (inhibition on the catalytic activity of nanozymes by antioxidant properties) make it challenging to achieve high specificity in detection. Therefore, the design of nanozymes with other catalytic properties or multiple reaction mechanisms to improve the selectivity of nanozymes for the detection of phytochemicals deserves further investigation.

• (2) The variety of components based on nanozymes detection in real samples is limited. In plant samples, complex compositions may affect the activity of nanozymes, leading to inaccurate results. Accordingly, detecting active ingredients in real samples often requires complex pretreatments. Developing nanozymes with specific adsorption or combination with other techniques (e.g., MIP) shows greater prospects.

• (3) In the detection of hazardous substances (e.g., heavy metal ions and pesticide residues), most samples are spiked with them rather than detecting their actual contents directly. This may be due to the low level of hazardous substances in the original sample and the lack of sensitivity of the assay. Modification of nanomaterials with small molecules (e.g., fluorescein derivatives and vitamin B6) that possess POD-like activity to enhance their enzyme-like activity or enable multiple enzyme activities are anticipated to improve the sensitivity of the assay.

• (4) Nanozyme-based methods for the detection of mycotoxins are mostly combined with immune methods or aptamers, making them more complex than direct colorimetric or fluorescent assays. Only one study was found on the direct colorimetric detection of AFB1 in corn and peanut samples, but its sensitivity is relatively low [89]. Hence, it is necessary to develop more sensitive and direct methods for easier analysis.

• (5) Currently, there is less literature on detecting heavy metal ions in plant samples (especially in CHMs) based on nanozymes. CHMs have been extensively used in disease treatment and healthcare for their unique therapeutic effects [144]. However, they are susceptible to contamination by chemicals in the environment. Therefore, evaluating the safety of CHMs using nanozymes is very important.

• (6) Although many nanozymes can mimic the catalysis activity of natural enzymes, their specificity is still inadequate and requires further optimization. In addition, some soluble transition metal nanozymes may be highly toxic and contaminate environment by releasing toxic metal ions. Consequently, it is important to develop nanozymes with improved specificity and biocompatibility.

Conclusions

This paper summarizes the applications of nanomaterials with enzyme-like activity in plant samples analysis from 2015 to the present, including the analysis of phytochemicals, organophosphorus pesticides, heavy metal ions, and mycotoxins. Improving the selectivity is a research priority for the detection of phytochemicals, which may be achieved through multimode detection and molecular imprinting. Furthermore, due to the trace levels of contaminates, it is crucial to improve sensitivity for detecting hazardous substances. There are various methods have been reported to achieve this goal, including the cascade reaction of natural enzymes with nanozymes and the enhancement of nanozymes’ catalytic activity through doping with other elements or material modification. In general, designing nanozymes with more enzyme-like activities and improving the specificity and sensitivity in their applications are the focus of future research.