2.1.2. Label-Based Optical Immunosensors

Jiang et al. [45] presented a paper-based microfluidic device (DON-Chip) for DON determination. In the competitive immunoassay, AuNPs were used for labelling. For signal reading, a low-powered digital microscope connecting to a computer's USB port was used for image acquisition and signal analysis to enable on-site determination. Detection of DON in aqueous extracts of food and feed was carried out by DON-chip, and the results were compared by those obtained by commercial DON ELISA, which showed linear correlation. The LOD of DON was 4.7 ng/g, and the linear working range was between 0.01–20 µg/g. For simultaneous determination of ZON and DON, Jin et al. [46] developed a novel dual near-infrared fluorescence-based lateral flow immunosensor (NIR-based LFIA). On the nitrocellulose membrane, DON and ZON conjugated to BSA were immobilised in the same test line. The anti-ZON and anti-DON antibodies were labelled by near-infrared dyes with distinct fluorescence characteristics as detection reagents. With the optimised sensor, the ZON and DON content of maize samples were determined with a LOD of 0.55 µg/kg and 3.8 µg/kg, respectively. The assay took 20 min to perform, providing a fast and sensitive tool for simultaneous determination of two mycotoxins (Figure 2). *Toxins* **2021**, *13*, x FOR PEER REVIEW 8 of 18

**Figure 2.** Operating principles of label-based optical immunosensors. (**A**) Paper-based microfluidic device; (**B**) microfluidic capillary chip. **Figure 2.** Operating principles of label-based optical immunosensors. (**A**) Paper-based microfluidic device; (**B**) microfluidic capillary chip.

In the electrochemical biosensors, the reaction between the target molecule and the

An electrochemical immunosensor to determine ZON in maize using modified screen-printed carbon electrodes (SPCE) was developed by Riberi et al. [60]. On the surface of the SPCE modified with multi-walled carbon nanotubes/polyethyleneimine dispersions and AuNPs, ZON polyclonal antibodies were immobilised. A competitive immunoassay was used for ZON determination where ZON presented in the sample, and a horseradish peroxidase (HRP)-labelled ZON conjugate competed for the limited amount of polyclonal antibodies immobilised on the surface. After that, hydrogen peroxide was added, and a steady-state current was obtained, which was proportional to the amount of ZON in the samples and was detected at a potential of −0.3V by amperometry. The biosensors showed good stability during at least four days. The calibration curve was linear

mulation, which can be quantified by voltammetric, potentiometric, amperometric, or conductometric techniques [49] (Figure 3). The use of electrochemical biosensors is very common due to their high sensitivity, selectivity, low cost, simplicity, and in some cases their miniaturisation, portability, and integration into automated devices [50–52]. In the last decade, the use of screen-printed electrodes (SPE) in electrochemical biosensor development has received great attention because they can be made of different materials and shapes and can be modified with a wide variety of nanomaterials, such as carbon nanotubes, graphene, and metallic nanoparticles as gold, silver, and magnetic nanoparticles coupled with different biological recognition elements (DNA, RNA, aptamers, enzymes,

in the ZON concentration range from 0.1 to 100 pg/mL.

*2.2. Electrochemical Immunosensors* 

antibodies) [53–59] (Figure 3A,B).

A multiplexed microfluidic capillary chip with smartphone detection for DON, OTA, and AFB1 determination was demonstrated [47]. A competitive immunoassay format was used to detect mycotoxins simultaneously, where mycotoxin-BSA conjugates were immobilised on a polydimethylsiloxane (PDMS) surface. Toxins present in the sample compete with the toxins immobilised on the surface for the binding site of the polyclonal antibodies conjugated with horseradish peroxidase. After that, hydrogen peroxide as a substrate and tetramethylbenzidine (TMB) as a chromophore were added, and the colorimetric signal was detected by a smartphone and analysed in ImageJ software. The assay could be performed in less than 10 min with a LOD of 10 ng/mL for DON, making the assay capable of fast, on-site analysis. Another smartphone-based sensor was developed by Liu et al. [48] using a dual fluorescence or colour detection mode device integrated with two lateral flow immunoassays for multiplex mycotoxin (DON, ZON) determination in cereals. When fluorescence detection was applied, the assays were more sensitive, but recoveries from maize for both formats were the same.

#### *2.2. Electrochemical Immunosensors*

In the electrochemical biosensors, the reaction between the target molecule and the recognition element by using electrochemical dyes or enzymatic reactions generates changes in the signal for conductance or impedance, measurable current, or change accumulation, which can be quantified by voltammetric, potentiometric, amperometric, or conductometric techniques [49] (Figure 3). The use of electrochemical biosensors is very common due to their high sensitivity, selectivity, low cost, simplicity, and in some cases their miniaturisation, portability, and integration into automated devices [50–52]. In the last decade, the use of screen-printed electrodes (SPE) in electrochemical biosensor development has received great attention because they can be made of different materials and shapes and can be modified with a wide variety of nanomaterials, such as carbon nanotubes, graphene, and metallic nanoparticles as gold, silver, and magnetic nanoparticles coupled with different biological recognition elements (DNA, RNA, aptamers, enzymes, antibodies) [53–59] (Figure 3A,B). *Toxins* **2021**, *13*, x FOR PEER REVIEW 9 of 18

**Figure 3.** Structure principles of electrochemical sensors. (**A**) Gold nanoparticles (AuNPs); (**B**) nanorods, nanotubes (Au, C, etc.); (**C**) molecular imprinting polymers (MIP); (**D**) aptamers. **Figure 3.** Structure principles of electrochemical sensors. (**A**) Gold nanoparticles (AuNPs); (**B**) nanorods, nanotubes (Au, C, etc.); (**C**) molecular imprinting polymers (MIP); (**D**) aptamers.

A differential pulse voltammetry (DPV) detection-based immunosensor using disposable SPE was prepared for ZON determination by Goud et al. [61]. On the activated sensor surface, a ZON-BSA conjugate was immobilised by the EDC/NHS method. A competitive assay format was used for ZON determination, and alkaline phosphatase-labelled antibody and 1-naphthyl phosphate (1-NP) as a substrate was used to detect primary antibody binding to the surface. The produced 1-naphthol was detected via DPV, which allowed the determination of the ZON concentration of the sample. The LOD was 0.25 An electrochemical immunosensor to determine ZON in maize using modified screenprinted carbon electrodes (SPCE) was developed by Riberi et al. [60]. On the surface of the SPCE modified with multi-walled carbon nanotubes/polyethyleneimine dispersions and AuNPs, ZON polyclonal antibodies were immobilised. A competitive immunoassay was used for ZON determination where ZON presented in the sample, and a horseradish peroxidase (HRP)-labelled ZON conjugate competed for the limited amount of polyclonal

A mesoporous silica-modified SPCE-based immunosensor was presented by Regiart

An electrochemical immunosensor fabricated on indium tin oxide (ITO)-coated glass was introduced by Lu et al. [63] for multiple mycotoxin determination. A dual-channel three-electrode sensor consisted of two working electrodes that were modified with AuNPs and functionalised with anti-FB1 and anti-DON antibodies and a Ag/AgCl pseudo-reference electrode etched on the ITO-coated glass and was integrated with a microfluidic channel. The binding of the toxin present in the sample to the antibody immobilized on the working electrode produced an electrochemical signal, which was detected by DPV. With this immunosensor set-up, a LOD of 97 pg/mL and 35 pg/mL could be achieved, and linear ranges of detection were 0.3–140 ng/mL and 0.2–60 ng/mL for FB1

The statistical parameters of the measurements, the cross reactivity, and the matrix analysed of electrochemical immunosensors for DON and ZON detection are summa-

was recognised and bound to the immobilised antibodies on the surface of the electrode. Then, to detect immunocomplex formation, HRP-conjugated anti-ZON antibodies were added, and hydrogen peroxide with 4-tert-butylcatechol (4-TBC) were used in a substrate and chromophore solution. The HRP enzyme catalyzes the oxidation of 4-TBC to 4-tertbutylbenzoquinone. The enzymatic product was detected by amperometry at −100 mV. The measured current was proportional to the concentration of ZON present in the sample. The linear measuring range of ZON detection was 1.88–45 ng/mL, and the LOD was

**Table 2.** Statistics of measuring parameters, cross reactivity, and the matrix analysed of electrochemical immunosensors

0.57 ng/mL in *Amaranthus cruentus* seeds.

and DON, respectively.

rized in Table 2.

for DON and ZON detection.

ng/mL, and the dynamic measuring range of ZON was 0.25–256 ng/mL.

antibodies immobilised on the surface. After that, hydrogen peroxide was added, and a steady-state current was obtained, which was proportional to the amount of ZON in the samples and was detected at a potential of −0.3V by amperometry. The biosensors showed good stability during at least four days. The calibration curve was linear in the ZON concentration range from 0.1 to 100 pg/mL.

A differential pulse voltammetry (DPV) detection-based immunosensor using disposable SPE was prepared for ZON determination by Goud et al. [61]. On the activated sensor surface, a ZON-BSA conjugate was immobilised by the EDC/NHS method. A competitive assay format was used for ZON determination, and alkaline phosphatase-labelled antibody and 1-naphthyl phosphate (1-NP) as a substrate was used to detect primary antibody binding to the surface. The produced 1-naphthol was detected via DPV, which allowed the determination of the ZON concentration of the sample. The LOD was 0.25 ng/mL, and the dynamic measuring range of ZON was 0.25–256 ng/mL.

A mesoporous silica-modified SPCE-based immunosensor was presented by Regiart et al. [62]. For the immunosensor anti-ZON antibodies were immobilised by GA on the surface of the modified electrode. During measurement, ZON presented in the sample was recognised and bound to the immobilised antibodies on the surface of the electrode. Then, to detect immunocomplex formation, HRP-conjugated anti-ZON antibodies were added, and hydrogen peroxide with 4-tert-butylcatechol (4-TBC) were used in a substrate and chromophore solution. The HRP enzyme catalyzes the oxidation of 4-TBC to 4-tert-butylbenzoquinone. The enzymatic product was detected by amperometry at −100 mV. The measured current was proportional to the concentration of ZON present in the sample. The linear measuring range of ZON detection was 1.88–45 ng/mL, and the LOD was 0.57 ng/mL in *Amaranthus cruentus* seeds.

An electrochemical immunosensor fabricated on indium tin oxide (ITO)-coated glass was introduced by Lu et al. [63] for multiple mycotoxin determination. A dual-channel three-electrode sensor consisted of two working electrodes that were modified with AuNPs and functionalised with anti-FB1 and anti-DON antibodies and a Ag/AgCl pseudoreference electrode etched on the ITO-coated glass and was integrated with a microfluidic channel. The binding of the toxin present in the sample to the antibody immobilized on the working electrode produced an electrochemical signal, which was detected by DPV. With this immunosensor set-up, a LOD of 97 pg/mL and 35 pg/mL could be achieved, and linear ranges of detection were 0.3–140 ng/mL and 0.2–60 ng/mL for FB1 and DON, respectively.

The statistical parameters of the measurements, the cross reactivity, and the matrix analysed of electrochemical immunosensors for DON and ZON detection are summarized in Table 2.


**Table 2.** Statistics of measuring parameters, cross reactivity, and the matrix analysed of electrochemical immunosensors for DON and ZON detection.

Deoxynivalenol (DON), Zearalenone (ZON), Fumonisin B1 (FB1), Aflatoxin B1 (AFB1), Aflatoxin M1 (AFM1), Ochratoxin A (OTA), Ochratoxin B (OTB), differential pulse voltammetry (DPV), no data (n.d.)

FB1

#### *2.3. Piezoelectric Immunosensors 2.3. Piezoelectric Immunosensors*

ZON DPV 0.25–256 ng/mL 0.25 ng/mL spiked beer,

ZON Amperometry 1.88–45 ng/mL 0.57 ng/mL *Amaranthus* 

0.2–60 ng/mL

Ochratoxin B (OTB), differential pulse voltammetry (DPV), no data (n.d.)

DON DPV 0.3–140 ng/mL

Quartz crystal microbalance (QCM) is a piezoelectric effect-based mass measuring system. The QCM sensor is made of a quartz crystal disk cut to a specific orientation with respect to the crystal axes and sandwiched between two metal electrodes (usually gold) that can be made to oscillate at a defined frequency by applying alternating voltage. Its resonant frequency depends on the thickness of the crystal (Figure 4). The thinner the applied crystal, the higher its resonant frequency and sensitivity. QCM monitors the mass or thickness of the adlayers on the surface of the quartz crystal. The main advantages of QCM are high sensitivity, high stability, fast response, and low cost. It also provides label-free detection capabilities for biosensor applications. However, QCM faces some disadvantages, as its performance significantly depends on the temperature and other environmental parameters, and its sensitivity falls short of the requirements when measuring low molecular weight substances [64]. In order to fulfill the requirements of high sensitivity regarding mycotoxin detection (as they are low molecular weight compounds, so they cannot generate sufficient frequency changes) piezoelectric biosensors need to apply competitive inhibition immunoassay formats, or the signal has to be amplified by applying secondary antibodies or nanoparticles. Quartz crystal microbalance (QCM) is a piezoelectric effect-based mass measuring system. The QCM sensor is made of a quartz crystal disk cut to a specific orientation with respect to the crystal axes and sandwiched between two metal electrodes (usually gold) that can be made to oscillate at a defined frequency by applying alternating voltage. Its resonant frequency depends on the thickness of the crystal (Figure 4). The thinner the applied crystal, the higher its resonant frequency and sensitivity. QCM monitors the mass or thickness of the adlayers on the surface of the quartz crystal. The main advantages of QCM are high sensitivity, high stability, fast response, and low cost. It also provides labelfree detection capabilities for biosensor applications. However, QCM faces some disadvantages, as its performance significantly depends on the temperature and other environmental parameters, and its sensitivity falls short of the requirements when measuring low molecular weight substances [64]. In order to fulfill the requirements of high sensitivity regarding mycotoxin detection (as they are low molecular weight compounds, so they cannot generate sufficient frequency changes) piezoelectric biosensors need to apply competitive inhibition immunoassay formats, or the signal has to be amplified by applying secondary antibodies or nanoparticles.

wine

spiked corn

AFB1 AFM1 85–90% OTA OTB

*cruentus* seeds n.d. [62]

sample n.d. [63]

[61]

*Toxins* **2021**, *13*, x FOR PEER REVIEW 10 of 18

**Mycotoxin Method Detection Range LOD Matrix Selectivity Reference**  ZON Amperometry 0.1 to 100 pg/mL 0.15 pg/mL spiked maize n.d. [60]

> 97 pg/mL 35 pg/mL

**Figure 4.** Structure principle of piezoelectric immunosensors. **Figure 4.** Structure principle of piezoelectric immunosensors.

Although there are several examples of piezoelectric immunosensors for mycotoxin determination in the recent scientific literature [1,65–69], there have been very few developments for the piezoelectric determination of ZON and DON. Very recently a portable, label-free QCM immunosensor was introduced by Liu et al. [70] for ZON determination in different food matrices. In the sensor, ZON-ovalbumin conjugate was immobilised with EDC/NHS on the surface of the mercaptodecylic acid-modified chip. The frequency response caused by the specific binding of anti-ZON antibody (100 µg/mL) on the chip surface was detected in the presence or absence of ZON. A high sensitivity of ZON determination with a LOD as low as 0.37 ng/mL was obtained, with excellent selectivity and stability. The effectiveness of the sensor was verified in spiked corn, wheat flour, soy sauce, and milk samples, and satisfactory recoveries were attained. The sensor could be reused Although there are several examples of piezoelectric immunosensors for mycotoxin determination in the recent scientific literature [1,65–69], there have been very few developments for the piezoelectric determination of ZON and DON. Very recently a portable, label-free QCM immunosensor was introduced by Liu et al. [70] for ZON determination in different food matrices. In the sensor, ZON-ovalbumin conjugate was immobilised with EDC/NHS on the surface of the mercaptodecylic acid-modified chip. The frequency response caused by the specific binding of anti-ZON antibody (100 µg/mL) on the chip surface was detected in the presence or absence of ZON. A high sensitivity of ZON determination with a LOD as low as 0.37 ng/mL was obtained, with excellent selectivity and stability. The effectiveness of the sensor was verified in spiked corn, wheat flour, soy sauce, and milk samples, and satisfactory recoveries were attained. The sensor could be reused six times without any significant attenuation of frequency of the sensor chip (below 10%) and could be stored for fifteen days without significant signal loss. The sensor allowed quick ZON determination since it took five minutes to measure a sample.

Nolan et al. [71] developed a mass-sensitive microarray biosensor working under the same principle as QCM for multiplex mycotoxin determination. The sensor consisted of 4x16 mass-sensitive transducer pixels. Each pixel consisted of a zinc oxide piezoelectric layer sandwiched between two electrodes where the top electrode was coated with silicon dioxide with a thin gold layer on the top where mycotoxin conjugates were immobilized, and the entire set-up was mounted on the top of an acoustic mirror. With the optimised sensor, simultaneous determination of T2-toxin, ZON, and FB1 were examined. To assess sensitivity, IC<sup>50</sup> values were calculated. Sensitivity of the multiplex assay were 6.1 ng/mL,

3.6 ng/mL, and 2.4 ng/mL, and the working range of the assay for T2, FB1, and ZON were 1.5–24.4 ng/mL, 0.9–14.3 ng/mL, and 0.6–9.6 ng/mL, respectively.

#### **3. Sensors Based on Artificial Recognition Elements**

MIPs are synthetic polymers that can be used to form an artificial receptor for the target analyte. They are synthesised by polymerisation of a monomer with a cross-linking agent in the presence of the target analyte. Upon cross-linking, a cavity is formed around the template, and after its removal, a recognition site appears for the target analyte. The formed polymer can be used as a recognition element in affinity-based sensors. MIPs are cheaper, have higher reusability, and are more resistant to pH and to ionic strength compared to antibodies; therefore, their use in sensor development is beneficial [72,73]. Aptamers are single-stranded nucleic acid (DNA or RNA) molecules with a high affinity to the target molecule. They are fabricated by an in vitro selection and amplification technology (SELEX) [74]. During several selection rounds, only those oligonucleotides are selected and enriched from the huge oligonucleotide library, which can bind with very high affinity to the specific molecular target. It can be stated that the affinity of aptamers can be as good as those of antibodies and in some cases, even better. In addition to that, aptamers are more stable and flexible and can be chemically modified, allowing their immobilisation in sensors.

#### *3.1. Aptasensors*

The use of aptamers over antibodies has been an emerging trend in the field of biosensor development in the last decades. Aptamers are synthetic, short, single-stranded nucleic acids with a high affinity to the target molecule. Due to their small size, high affinity, high stability, and specificity, they offer many advantages over conventional antibodies as recognition elements. Having such high affinity, aptamer-based homogeneous and heterogeneous sensors have emerged as a promising tool among the biosensors (Figure 3D). Fluorescent, colorimetric, and electrochemical detection methods are commonly used in these sensor systems. A fluorometric aptamer-based method was developed for simultaneous determination of ZON and FB1 using gold nanorods (AuNRs) and upconversion nanoparticles (UCNPs) [75]. In the sensor, UCNPs were modified with aptamers for ZON and FB1. The functionalised UCNPs were attached with their corresponding complementary nucleic acid (cDNA) sequences. To the AuNPs, different cDNAs for ZON and FB1 were attached, and the AuNPs and the UCNPs were assembled together. In the presence of ZON and FB1 in the sample, the biocomplex of UCNPs-AuNRs will be unstable, and the UCNP part separates from the complex, resulting in the recovery of fluorescence signals. Under 980-nm laser excitation, ZON was detected at 606 nm and FB1 at 753 nm. The LODs of the assay for ZON and FB1 were 1 pg/mL and 3 fg/mL, respectively, with average recoveries from spiked maize samples of 90 to 107%.

Similarly, a fluorescent aptasensor created through UCNPs was presented for ZON determination in corn and beer [76]. A ZON-specific aptamer was used as a recognition probe, while the complementary strand was adopted as a signal probe. In the sensor, ZON aptamer was immobilised on the surface of the amino-modified magnetic nanoparticles, while cDNA was immobilised on the surface of UCNPs and were mixed together to form the duplex structure. When ZON is present in the sample, the ZON-aptamer dissociates from the complex and binds to ZON; therefore, a decrease in the fluorescence intensity occurs. For excitation, a 980 nm laser light was used, and ZON was detected at 543 nm. In this sensing platform, a linear response of 0.05–100 ng/mL was obtained between the fluorescence signal and ZON levels with a LOD of 0.126 µg/kg in corn and 0.007 ng/mL for beer, demonstrating that the developed aptasensor offered a novel approach for ZON analysis in food. Li et al. [77] presented an aptasensor for ZON determination in maize samples that was based on fluorescence resonance energy transfer (FRET) between fluorescent UCNPs modified with aptamer as donors and graphene oxide modified with carboxyl groups as acceptor. When UCNPs and functionalised graphene oxide were at a close

distance (less than 10 nm), fluorescence quenching was noticed. As the aptamers prefer to bind to their corresponding mycotoxins, in the presence of ZON, the formation of aptamers change, so aptamer modified-UCNPs are far away from the surface of the functionalised graphene oxide. The presented sensor had a wide working range (0.005–100 ng/mL), good stability (28 days), and the results showed that the aptamer-UCNP-functionalised graphene oxide probe provided a rapid, accurate, and simple to use system for ZON detection.

Azri et al. [78] fabricated an electrochemical label-free competitive aptasensor for ZON determination. The sensor had a working range of 0.01 to 1000 ng/mL ZON concentration with a LOD of 0.017 ng/mL. With the established aptasensor, ZON concentrations of maize grain extracts were determined. For ZON determination, He et al. [79] described a voltammetric aptasensor based on the use of porous platinum nanotubes/AuNPs and thionine-labelled graphene oxide for signal amplification. The working range of the aptasensor was 0.5 pg/mL to 0.5 µg/mL for ZON with a LOD of 0.17 pg/mL.

Recently, an aptasensor for ultrasensitive detection of ZON by using CoSe<sup>2</sup> nanocrystal /AuNRs, 3D structured DNA-PtNi@Co-metal-organic framework networks, and nicking enzyme as signal amplification system was proposed [80]. In the sensor DPV detection method was used for ZON determination. Comparing to other ZON methods, the aptasensor possessed outstanding sensitivity (LOD = 1.37 fg/mL) and wider linear range (10.0 fg/mL to 10.0 ng/mL). In addition, no additional substrate was needed compared to conventional enzymatic amplification by substrate cycling. Ong et al. [81] described a novel aptasensor for DON determination where they used iron nanoflorets graphene nickel (INFGN) as a transducer. The INFGN enabled a feasible bio-capturing due to its large surface area where the hydroxyl groups act as linkers. The biomolecular interaction in the sensor results in conductivity changes determined by current-voltage measurement using a picoammeter. The sensor showed good stability, it retained 30.65% of its activity after 48 h, and provided highly sensitive and selective detection of DON at a LOD of 2.11 pg/mL. Another research group used the 3D sakura-shaped copper (II) ions@L-glutamic acid nano-metal-organic coordination polymers (MOCPs) for the first time to develop an electrochemical aptasensor for ultrasensitive detection of ZON. Cronoamperometry was used for ZON determination. Under optimal conditions, dynamic range of 1 fg/mL to 100 ng/mL ZON was obtained with a LOD of 0.45 fg/mL [82].

Han et al. [83] presented a co-reduced molybdenum disulphide and gold nanoparticles (rMoS2-Au)-based electrochemical aptasensor for ZON and FB1 simultaneous detection. For sensor fabrication on the surface of the reduced molybdenum disulphide and AuNPs, coated glassy carbon electrode ZON and FB1 aptamers were conjugated. The corresponding cDNA sequences and thionine and 6-(ferrocenyl)hexanethiol as probes for ZON and FB1 detection were immobilised on AuNPs, which were bound to the aptamers through the complementary base pairing. In the presence of ZON and FB1, the labelled corresponding cDNAs are replaced by the target molecule, resulting in signals proportional to the concentrations of the analytes. Differential pulse voltammetry was used to detect the concentrations of the mycotoxins. The aptasensor allowed ZON and FB1 determination in the range of 1×10−3–10 ng/mL and 1×10−3–1×10<sup>2</sup> ng/mL, respectively. The sensor possesses the LOD of 5<sup>×</sup> <sup>10</sup>−<sup>4</sup> ng/mL. The performance of the aptasensor was successfully demonstrated in real maize samples with satisfactory recoveries.

The statistical parameters of the measurements, the cross reactivity, and the matrix analysed of aptasensors for DON and ZON detection are summarized in Table 3.


**Table 3.** Statistics of measuring parameters, cross reactivity, and the matrix analysed of aptasensors for DON and ZON detection.

Deoxynivalenol (DON), Zearalenone (ZON), α-zearalenol (α-ZEL), β-zearalenol (β-ZEL), Zearalenone-14-Glucoside (ZON-14-Glc), Ochratoxin A (OTA), Ochratoxin B (OTB), Aflatoxin B1 (AFB1), Aflatoxin B2 (AFB2), Aflatoxin M1 (AFM1), Fumonisin B1 (FB1), Fumonisin B2 (FB2), Patulin (PAT), differential pulse voltammetry (DPV), signal is not significant (n.s.), no data (n.d.).

#### *3.2. Molecularly Imprinted Polymer Sensors*

In recent years, MIPs are widely used primarily in the SPR biosensor technique. Comparing to antibodies, MIPs are more resistant to harsh regeneration conditions and are less likely to lose their binding capability. Although there are several methods to prepare MIPs for sensor applications, the most common method is the in situ polymerisation directly onto the sensor surface (Figure 3C). Choi et al. [84] developed an SPR sensor for ZON determination using MIPs as recognition elements. On the gold sensor surface, a molecularly imprinted polypyrrole film was prepared by electropolymerisation in the presence of ZON as a template. The sensor had a linear response in the range of 0.3–3000 ng/mL for ZON, and the LOD was 0.3 ng/g in corn samples. They also prepared a similar MIP-based SPR sensor for the determination of DON in which the linear measuring range was between 0.1–100 ng/mL. The selectivity of the MIP layer for 3- and 15-acetyl-DON was found to be 19% and 44%, respectively [85].

Sergeyeva et al. [86] developed a novel sensor for ZON detection in cereals suitable for field application. A ZON-selective urethane-acrylate MIP membrane was used to form the sensor, and the natural fluorescence of ZON was analysed by a Spotxel®Reader smartphone application. In the direct sensing mode, the LOD of ZON was 126 µg/kg, but the competitive sensing mode allowed a sensitivity improvement to a LOD of 1.26 µg/kg.

## **4. Conclusions**

Quick, easy to use, and sensitive determination of mycotoxins are extremely important in the food and feed industry because the use of mycotoxin-contaminated commodities poses health risks to the consumers and to livestock as well. The application of biosensors could be an expedient alternative over advanced instrumental chromatographic techniques, as they offer cost-effective, rapid, portable, on-site determination possibilities of mycotoxins. Although developments of several immunosensors for mycotoxin determination have been reported in the scientific literature, they are mainly focused on aflatoxin and ochratoxin as target analytes, but much less attention has been paid to the determination of ZON and DON, and the reports dealing with masked mycotoxins are unduly rare. For the detection of small molecular mass analytes, substantial advances have occurred in

the fields of electrochemical and optical immunosensing. Efforts for both types of these sensors are aimed to improve biosensor characteristics, including sensitivity, selectivity, fast response, and low cost; therefore, incorporation of nanomaterials (nanoparticles, nanorods, nanotubes, nanowires) into biosensors are being widely studied. The advantages of using nanoparticles are that they either increase the sensor surface area suitable for biomolecule immobilisation or enhance the signal derived from the immunocomplex formation. It has been found that nanomaterials applied in biosensors as signal amplification tags can improve sensitivity and can reduce the LOD by several orders of magnitude. The use of the favourable properties of nanomaterials in the determination of mycotoxins via immunosensors is particularly important, as these analytes are low molecular weight substances; therefore, their detection is challenging. Another emerging trend in biosensor development is the application of aptamers and MIPs as synthetic receptors in biosensor fabrication. During the past decade, the focus of the attention has turned towards the development of aptasensors due to the stability, selectivity, and sensitivity of these oligonucleotide-type artificial recognition elements. Despite new achievements, areas demanding more research still exist, particularly in the fields of masked mycotoxins and multiplex mycotoxin determination.

**Author Contributions:** All the authors participated in the preparation of the manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Hungarian National Research, Development, and Innovation Office, projects TKP2020-NKA 24 "Tématerületi kiválóság program" in the Thematic Excellence Programme 2020–2021, and NVKP\_16-1-2016-0049 "In situ, complex water quality monitoring by using direct or immunofluorimetry and plasma spectroscopy".

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

