Enzyme Cascade Amplification-Based Immunoassay Using Alkaline Phosphatase-Linked Single-Chain Variable Fragment Fusion Tracer and MnO2 Nanosheets for Detection of Deoxynivalenol in Corn Samples

Deoxynivalenol (DON) is a common mycotoxin that contaminates cereals. Therefore, the development of sensitive and efficient detection methods for DON is essential to guarantee food safety and human health. In this study, an enzyme cascade amplification-based immunoassay (ECAIA) using a dual-functional alkaline phosphatase-linked single-chain fragment variable fusion tracer (scFv-ALP) and MnO2 nanosheets was established for DON detection. The scFv-ALP effectively catalyzes the hydrolysis of ascorbyl-2-phosphate (AAP) to produce ascorbic acid (AA). This AA subsequently interacts with MnO2 nanosheets to initiate a redox reaction that results in the loss of oxidizing properties of MnO2. In the absence of ALP, MnO2 nanosheets can oxidize 3,3′,5,5′-tetramethylbenzidine (TMB) to produce the blue oxidized product of TMB, which exhibits a signal at a wavelength of 650 nm for quantitative analysis. After optimization, the ECAIA had a limit of detection of 0.45 ng/mL and a linear range of 1.2–35.41 ng/mL. The ECAIA exhibited good accuracy in recovery experiments and high selectivity for DON. Moreover, the detection results of the actual corn samples correlated well with those from high-performance liquid chromatography. Overall, the proposed ECAIA based on the scFv-ALP and MnO2 nanosheets was demonstrated as a reliable tool for the detection of DON in corn samples.


Introduction
Deoxynivalenol (DON) is a secondary metabolite produced by Fusarium graminearum [1].Due to the prevalence of toxin-producing strains in warm and humid environments, DON contamination is widespread in stored crops, feed, and foodstuffs [2].Studies have shown that DON can inhibit animal protein synthesis, damage immune cells, severely impact the immune system, and potentially lead to cancer [3,4].To address the serious health risks associated with DON, many countries and organizations have established limits on its presence in food [5].For instance, the Codex Alimentarius Commission (CAC) has instituted DON limits of 2000 µg/kg for corn, barley, and wheat, and 1000 µg/kg for derivative cereal products.In China, the regulatory threshold for grains and their derivatives stands at 1000 µg/kg.DON exhibits excellent thermal stability, surviving high-temperature cooking and baking processes in contaminated food [6].Therefore, it Foods 2024, 13, 2009 2 of 12 is crucial to establish sensitive and efficient detection methods to ensure food safety and minimize economic losses.
Currently, instrumental analysis methods are widely used for the detection of exogenous contaminants such as mycotoxins, pesticides, and veterinary drugs [7][8][9].However, the complex sample preparation processes and high costs associated with these methods have spurred the development of immunoassay methods.Immunoassay methods mainly include enzyme-linked immunosorbent assay (ELISA), electrochemical immunoassay, chemiluminescent immunoassay, fluorescent immunoassay, lateral flow immunoassay, and microfluidic-based immunoassay [10][11][12][13][14][15][16][17][18][19].Although these immunoassays differ in their detection formats, they all rely on the specific binding between antigens and antibodies.Despite the common use of monoclonal antibodies in immunoassays, limitations in terms of high cost and complex preparation steps have hindered their further applications [20,21].
The emergence of genetically engineered recombinant antibodies overcomes the limitations of traditional intact antibodies [22][23][24].With the development of phage display technology, genetically engineered antibodies can be produced through simplified and quantitative biopanning, replacing the traditional complex hybridoma technology [21,25].
Research has shown that genetically engineered recombinant antibodies have been successfully expressed in prokaryotic and eukaryotic cells [26,27].Among them, the single-chain fragment variable (scFv) has been very attractive because of its small size and low production cost [25,28].Moreover, the significant advantage of the scFv is its ease of genetic manipulation [29], which contributes to the generation of many bifunctional fusion proteins [30,31].In this study, the scFv was fused with alkaline phosphatase (scFv-ALP) for expression, enabling the protein to possess both the antibody's specific recognition and the enzyme's catalytic amplification ability.Despite their high catalytic efficiency and specificity, protein-based conventional biocatalysts are often characterized by poor stability and susceptibility to environmental influences, thereby significantly increasing storage costs [32].
Advancements in nanotechnology have led to the exploration of "nanozymes" that simulate the catalytic functions of traditional biological enzymes [32,33].For instance, Yan's team first demonstrated the peroxidase-like activity of iron oxide (Fe 3 O 4 ) nanoparticles and applied it to immunoassays [34].Compared to biological enzymes, nanozymes are more stable, efficient, and capable of large-scale production [35].In simple terms, nanozymes can be classified into oxidoreductases and hydrolases based on their catalytic principles.For example, metal-based, carbon-based, and metal-organic framework nanozymes have been widely used in food safety testing [32].Among them, MnO 2 nanosheets are a type of metalbased nanozyme with peroxidase-like properties.Its unique two-dimensional (2D) structure provides a large surface area, enabling efficient catalytic performance [36].Similar to horseradish peroxidase (HRP), MnO 2 nanosheets can oxidize 3,3 ′ ,5,5 ′ -tetramethylbenzidine (TMB) to the blue oxidized product of TMB (oxTMB).Additionally, research has found that ascorbic acid or glutathione can reduce MnO 2 nanosheets to Mn 2+ [36,37].Therefore, the catalytic properties of nanozymes render them applicable in rapid detection for food safety.
In this study, to expand the application of the scFv-ALP in immunoassays, an enzyme cascade amplification immunoassay (ECAIA) for deoxynivalenol (DON) was developed using the scFv-ALP and MnO 2 nanosheets.The construction of the MnO 2 nanosheets and TMB-based (MnO 2 -TMB) sensing system and its feasibility for detecting the scFv-ALP were described in detail.By integrating the scFv-ALP-based immunoassay system with the MnO 2 -TMB sensing system, the ECAIA can implement the detection of DON in food.The optimization of experimental parameters and the methodological evaluation of ECAIA were described in detail.

Expression, Purification, and Characterization of the scFv-ALP Fusion Tracer
The engineered E. coli BL21(DE3) strain was inoculated into the culture medium for the auto-induction expression of the scFv-ALP fusion tracer as described previously [39].After auto-induction expression, the bacterial cells from 200 mL culture were collected by centrifugation and resuspended in 10 mL of Tris-buffered saline (TBS, 50 mM, pH 7.4) for sonication.After centrifugation, the supernatant was transferred onto a column (1.6 × 2.5 cm) preloaded with Ni-NTA agarose for purification and elution according to the manufacturer's manual.The eluted protein was dialyzed in TBS at 4 • C for 72 h and then quantified using a Micro Drop ultramicro UV spectrophotometer (BIO-DL Corporation, Shanghai, China).Further validation on the expression and purification of the fusion tracer was performed using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blot as described previously [40].The purified fusion tracer was aliquoted for frozen storage at −20 • C before use.
To validate the functionality of the fusion tracer, the activities of both the antibody and enzyme were tested.Firstly, the antibody activity was assessed by an indirect ELISA as follows.Briefly, a 96-well microplate was coated with 100 µL/well of DON-BSA (2 µg/mL in TBS) at 37 • C for 2 h and blocked with 300 µL/well of 3% (m/v) skim milk powder in TBS at 37 • C for 1 h.After washing three times using the TBS with 0.05% (v/v) Tween-20 (TBST), the microplate was incubated with 100 µL/well of the scFv-ALP with various concentrations at room temperature for 1 h.Then, 100 µL/well of HRP conjugated anti-his tag mouse monoclonal antibody (0.33 µg/mL in TBS) was added and incubated at 37 • C for 1 h.After washing four times with TBST, the microplate was incubated with 100 µL/well of TMB substrate solution for color reaction (37 • C, 10 min).The reaction was terminated by adding 50 µL/well of 2 M H 2 SO 4 .The absorbance at 450 nm (OD 450 ) of the yellow products was measured using a microplate reader (ST-360, Shanghai Kehua Bio-Engineering Co., Ltd., Shanghai, China).The ∆A value was calculated using the formula ∆A = (OD 450 of MP-ALP) − (OD 450 of the blank well).A curve of ∆A versus the scFv-ALP concentration was plotted.Secondly, the enzyme activity was evaluated using a similar procedure, except that the scFv-ALP was not incubated with the secondary antibody.Instead, 100 µL of 0.1% (w/v) pNPP-diethanolamine buffer was added to each well, and the plate was incubated in the dark for 10 min.The ∆A value was calculated using the formula ∆A = (OD 405 of scFv-ALP) − (OD 405 of the blank well).A curve of ∆A versus the scFv-ALP concentration was plotted.

Synthesis and Characterization of MnO 2 Nanosheets
The MnO 2 nanosheets were synthesized as described previously with minor modifications [41].Briefly, a mixture containing 20 mL of 0.6 M TMA•OH and 2 mL of 30% H 2 O 2 was added to 10 mL of 0.3 M MnCl 4 •solution.After vigorous stirring at room temperature for 12 h, the mixed solution was centrifugated (10,000× g, 10 min) to collect the precipitate.Subsequently, the precipitate was washed three times with ultrapure water and anhydrous ethanol, followed by drying at 60 • C. The dried precipitate was dispersed in ultrapure water by sonication for 16 h.The un-separated MnO 2 nanosheets were removed by centrifugation (600× g, 10 min), and the supernatant was transferred for another centrifugation (800× g, 10 min) before storage at 4 • C. The concentration of the as-prepared MnO 2 nanosheets in the aqueous solution was determined by the Lambert-Beer law, with a molar extinction coefficient of 9.6 × 10 3 M −1 cm −1 at 480 nm.The characterization of the MnO 2 nanosheets was performed using JEM 2100 transmission electron microscopy (TEM) (JEOL Ltd., Tokyo, Japan), Bruker Dimension Edge atomic force microscopy (AFM) (Bruker Corporation, Billerica, MA, USA), and Malvern Zetasizer Nano ZS90 (Malvern Instruments Ltd., Malvern, UK).The UV-visible absorption spectra and absorbance were obtained using a spectral scanning multi-mode reader SP-Max 3500FL (Flash Spectrum Inc., Shanghai, China).

Enzyme Cascade Amplification-Based Immunoassay for DON
Based on the dual-functional fusion protein scFv-ALP and MnO 2 nanosheets, an enzyme cascade amplification-based immunoassay (ECAIA) for detecting DON was constructed as follows.First, a 96-well microplate was pretreated and blocked as described in Section 2.2.Then, 50 µL of a series of DON standard solutions and 50 µL of the scFv-ALP (10 µg/mL in TBS) were added into the microplate and incubated at 37 • C for 1 h.After each incubation step, the microplate was washed three times with TBST.Subsequently, the microplate was incubated with 40 µL of 250 µM AAP, 10 µL of MnO 2 nanosheets solution (1 mM), and 100 µL of NaAc-Hac buffer (20 mM NaAc, 20 mM HAc, pH adjusted to 3.8) at 37 • C for 1 h.Finally, 100 µL/well of TMB substrate solution was added for incubation in the dark at 37 • C for 10 min, and the absorbance at 650 nm (OD 650 ) for each well was measured using a spectral scanning multi-mode reader SP-Max 3500FL (Flash Spectrum Inc., Shanghai, China).For the quantitative analysis of DON, a standard competitive inhibition curve was established by plotting the logarithm of DON concentration against the percentage binding rate using a four-parameter logistic equation of Origin2019 (OriginLab Corporation, Northampton, MA, USA).The percentage binding rate was calculated using the following formula: binding rate (%) = (A − A blank )/(A 0 − A blank ) × 100%, where A is the absorbance in the presence of DON and scFv-ALP, A blank is the absorbance of blank without DON and scFv-ALP, and A 0 is the absorbance in the presence of the scFv-ALP and without DON.The experimental results were analyzed and judged based on the half-maximal inhibitory concentration (IC 50 ).

Selectivity of the ECAIA for DON
To validate the selectivity of the developed ECAIA, four common cereal mycotoxins including AFB 1 , OTA, ZEN, and FB 1 were used to replace DON for cross-reaction analysis.The percentage cross-reactivity (CR) rate was calculated as follows: CR (%) = [IC 50 (substitute)/IC 50 (DON)] × 100%.

Sample Analysis and Validation
The corn samples stored in the lab were pretreated and analyzed as described below.First, 1 g of the ground cereal sample was accurately weighed into a 15 mL centrifugal tube with 4 mL of TBS.The sample was extracted by vigorously shaking on a horizontal shaker for 20 min.After centrifugation, the supernatant was transferred and diluted properly with TBS for ECAIA analysis.For the spike-and-recovery experiment, the negative samples validated by HPLC with an ultraviolet detector (HPLC-UVD) were spiked with various levels of DON standard, followed by extraction and measurement as described above.To further verify the effectiveness of the proposed method for DON, the authentic cereal samples were synchronously pretreated for analysis by HPLC-UVD (Table S1) according to the national standard GB5009.111-2016 of China [42] with minor modifications, as described in the Supplementary Materials.

Expression, Purification, and Characterization of the scFv-ALP
The bifunctional fusion protein scFv-ALP was obtained by auto-induced expression in the engineered E. coli BL21(DE3) strain containing the recombinant plasmid pET25b-scFv-ALP.The purified protein was characterized using SDS-PAGE.As shown in Figure 1A, a predicted single clear band appeared at around 81 kDa, which corresponds to the calculated molecular weight of the scFv-ALP based on its nucleotide sequence.Moreover, the result indicated that the target protein scFv-ALP with high purity was successfully prepared.Subsequently, we further investigated whether both scFv and ALP maintained their activity in the dual-functional fusion protein scFv-ALP.As shown in Figure 1B, the OD 405 and OD 450 values increased and reached a plateau as the concentration of the scFv-ALP fusion tracer increased from 0.16 µg/mL to 10 µg/mL.Therefore, after prokaryotic expression, both the scFv and ALP of the fusion protein remained active.Furthermore, the good detection performance of the scFv-ALP was validated through the direct competitive ELISA, and an IC 50 of 5.85 ng/mL was obtained from the standard competitive inhibition curve (Figure 1C).
negative samples validated by HPLC with an ultraviolet detector (HPLC-UVD) were spiked with various levels of DON standard, followed by extraction and measurement as described above.To further verify the effectiveness of the proposed method for DON, the authentic cereal samples were synchronously pretreated for analysis by HPLC-UVD (Table S1) according to the national standard GB5009.111-2016 of China [42] with minor modifications, as described in the Supplementary Materials.

Expression, Purification, and Characterization of the scFv-ALP
The bifunctional fusion protein scFv-ALP was obtained by auto-induced expression in the engineered E. coli BL21(DE3) strain containing the recombinant plasmid pET25b-scFv-ALP.The purified protein was characterized using SDS-PAGE.As shown in Figure 1A, a predicted single clear band appeared at around 81 kDa, which corresponds to the calculated molecular weight of the scFv-ALP based on its nucleotide sequence.Moreover, the result indicated that the target protein scFv-ALP with high purity was successfully prepared.Subsequently, we further investigated whether both scFv and ALP maintained their activity in the dual-functional fusion protein scFv-ALP.As shown in Figure 1B, the OD405 and OD450 values increased and reached a plateau as the concentration of the scFv-ALP fusion tracer increased from 0.16 µg/mL to 10 µg/mL.Therefore, after prokaryotic expression, both the scFv and ALP of the fusion protein remained active.Furthermore, the good detection performance of the scFv-ALP was validated through the direct competitive ELISA, and an IC50 of 5.85 ng/mL was obtained from the standard competitive inhibition curve (Figure 1C).

Characterization of MnO2 Nanosheets
Due to the unique structural features of MnO2 nanosheets, they can serve as a peroxidase-like catalyst to oxidize TMB and induce color change.In this study, the synthesized two-dimensional nanozyme-MnO2 nanosheets were characterized using TEM, AFM, and a Zeta potential analyzer.As shown in Figure 2A, the TEM image reveals a typical twodimensional sheet-like morphology with an average diameter of approximately 600-700 nm and obvious wrinkles on the surface.As shown in Figure 2B, the AFM result also indicates that the sample has an irregular sheet-like structure with a thickness of approximately 1.5 nm.Moreover, the MnO2 nanosheets are uniformly dispersed in the aqueous solution, which is beneficial for exposing the catalytic sites of the nanozyme and improving its catalytic reaction efficiency.Additionally, the Zeta potential test of the MnO2

Characterization of MnO 2 Nanosheets
Due to the unique structural features of MnO 2 nanosheets, they can serve as a peroxidase-like catalyst to oxidize TMB and induce color change.In this study, the synthesized two-dimensional nanozyme-MnO 2 nanosheets were characterized using TEM, AFM, and a Zeta potential analyzer.As shown in Figure 2A, the TEM image reveals a typical two-dimensional sheet-like morphology with an average diameter of approximately 600-700 nm and obvious wrinkles on the surface.As shown in Figure 2B, the AFM result also indicates that the sample has an irregular sheet-like structure with a thickness of approximately 1.5 nm.Moreover, the MnO 2 nanosheets are uniformly dispersed in the aqueous solution, which is beneficial for exposing the catalytic sites of the nanozyme and improving its catalytic reaction efficiency.Additionally, the Zeta potential test of the MnO 2 nanosheets shows a negative Zeta potential of −25 mV, indicating that the nanosheets carry negative charges on the surface (Figure 2C).These negative charges contribute to the good dispersion and stability of the MnO 2 nanosheets in water.In conclusion, the MnO 2 nanosheets synthesized in this study exhibit a two-dimensional sheet-like nanostructure, good stability, and morphological characteristics.The results of the TEM, AFM, and Zeta potential tests provide important insights into the structural and physicochemical properties of the MnO 2 nanosheets and demonstrate their potential as a highly effective enzyme mimic.
nanosheets shows a negative Zeta potential of −25 mV, indicating that the nanosheets carry negative charges on the surface (Figure 2C).These negative charges contribute to the good dispersion and stability of the MnO2 nanosheets in water.In conclusion, the MnO2 nanosheets synthesized in this study exhibit a two-dimensional sheet-like nanostructure, good stability, and morphological characteristics.The results of the TEM, AFM, and Zeta potential tests provide important insights into the structural and physicochemical properties of the MnO2 nanosheets and demonstrate their potential as a highly effective enzyme mimic.

Feasibility Analysis and Optimization of the Nanozyme Sensing System for the scFv-ALP
The principle is illustrated in Scheme 1, wherein ALP catalyzes the AAP and yields the reducing agent AA.AA, in turn, modulates the MnO2-TMB system.Specifically, MnO2 nanosheets catalyze oxidative reactions with both AA and TMB, thereby generating the detection signal.To validate the feasibility of the proposed nanozyme sensing system for detecting the scFv-ALP, a control experiment was conducted in which various combinations of reaction components within the system were tested.After the reaction, the absorbance spectra at 400-750 nm were scanned using a UV-Vis spectrophotometer, and the results are shown in Figure 3.The result indicated that in the presence of only one component (scFv-ALP, MnO2 nanosheets, or TMB solution), there was no significant change observed in the UV-Vis absorption spectra (curves a, b, and c).However, upon the coexistence of MnO2 nanosheets and TMB, the solution changed from colorless to blue, accompanied by the emergence of a prominent absorption peak at 650 nm (curve f and inset f).It was attributed to the peroxidase-like activity of MnO2 nanosheets, which can oxidize TMB to blue oxTMB.Furthermore, we mixed Mn 2+ with TMB solution to evaluate the oxidation of Mn 2+ to TMB.The result showed that there was no significant change in the absorption spectra of the system and the solution remained unchanged (inset e), indicating that Mn 2+ did not undergo an oxidation-reduction reaction with TMB.When AA or AAP was introduced to the system, the results depicted in curves g and h of Figure 3 revealed interesting patterns.Specifically, the system containing AAP exhibited a curve similar to that observed in curve f: the solution changed from colorless to blue, accompanied by the emergence of a prominent absorption peak at 650 nm.However, when AA was added to the solution, the result was quite different.Instead of the color change, the solution remained colorless.This is because reducing AA reduced the MnO2 nanosheets, destroying their oxidizing properties and leaving them unable to continue catalyzing the oxidation of TMB to cause a color change in the solution.At the same time, the partially reduced AAP could not completely reduce MnO2 nanosheets to Mn 2+ , and the remaining MnO2 nanosheets oxidized TMB, resulting in a lower OD650 nm than curve f (the color in inset h is slightly lighter than that in f).Based on the feasibility verification above, when the scFv-ALP, AAP, MnO2 nanosheets, and TMB are present in the system simultaneously, the dual-functional fusion protein scFv-ALP catalyzes the hydrolysis of AAP to generate AA, which interacts with the MnO2 nanosheets.As a result, the solution remains

Feasibility Analysis and Optimization of the Nanozyme Sensing System for the scFv-ALP
The principle is illustrated in Scheme 1, wherein ALP catalyzes the AAP and yields the reducing agent AA.AA, in turn, modulates the MnO 2 -TMB system.Specifically, MnO 2 nanosheets catalyze oxidative reactions with both AA and TMB, thereby generating the detection signal.To validate the feasibility of the proposed nanozyme sensing system for detecting the scFv-ALP, a control experiment was conducted in which various combinations of reaction components within the system were tested.After the reaction, the absorbance spectra at 400-750 nm were scanned using a UV-Vis spectrophotometer, and the results are shown in Figure 3.The result indicated that in the presence of only one component (scFv-ALP, MnO 2 nanosheets, or TMB solution), there was no significant change observed in the UV-Vis absorption spectra (curves a, b, and c).However, upon the coexistence of MnO 2 nanosheets and TMB, the solution changed from colorless to blue, accompanied by the emergence of a prominent absorption peak at 650 nm (curve f and inset f).It was attributed to the peroxidase-like activity of MnO 2 nanosheets, which can oxidize TMB to blue oxTMB.Furthermore, we mixed Mn 2+ with TMB solution to evaluate the oxidation of Mn 2+ to TMB.The result showed that there was no significant change in the absorption spectra of the system and the solution remained unchanged (inset e), indicating that Mn 2+ did not undergo an oxidation-reduction reaction with TMB.When AA or AAP was introduced to the system, the results depicted in curves g and h of Figure 3 revealed interesting patterns.Specifically, the system containing AAP exhibited a curve similar to that observed in curve f: the solution changed from colorless to blue, accompanied by the emergence of a prominent absorption peak at 650 nm.However, when AA was added to the solution, the result was quite different.Instead of the color change, the solution remained colorless.This is because reducing AA reduced the MnO 2 nanosheets, destroying their oxidizing properties and leaving them unable to continue catalyzing the oxidation of TMB to cause a color change in the solution.At the same time, the partially reduced AAP could not completely reduce MnO 2 nanosheets to Mn 2+ , and the remaining MnO 2 nanosheets oxidized TMB, resulting in a lower OD 650 nm than curve f (the color in inset h is slightly lighter than that in f).Based on the feasibility verification above, when the scFv-ALP, AAP, MnO 2 nanosheets, and TMB are present in the system simultaneously, the dual-functional fusion protein scFv-ALP catalyzes the hydrolysis of AAP to generate AA, which interacts with the MnO 2 nanosheets.As a result, the solution remains colorless and does not exhibit an absorption peak at 650 nm (curve and inset i).Therefore, the scFv-ALP and MnO 2 -TMB system can be used to develop an enzyme cascade amplification immunoassay for detecting DON.On this basis, to obtain the optimal sensing performance of the MnO2-TMB system, further optimization was performed to determine the optimal concentration of MnO 2 nanosheets (1.4 mM) and AAP (250 µM), and the optical incubation time of 120 min for color development, as shown in Figure S1 and the Supplementary Material.
the scFv-ALP and MnO2-TMB system can be used to develop an enzyme cascade amplification immunoassay for detecting DON.On this basis, to obtain the optimal sensing performance of the MnO2-TMB system, further optimization was performed to determine the optimal concentration of MnO2 nanosheets (1.4 mM) and AAP (250 µM), and the optical incubation time of 120 min for color development, as shown in Figure S1 and the Supplementary Material.the scFv-ALP and MnO2-TMB system can be used to develop an enzyme cascade amplification immunoassay for detecting DON.On this basis, to obtain the optimal sensing performance of the MnO2-TMB system, further optimization was performed to determine the optimal concentration of MnO2 nanosheets (1.4 mM) and AAP (250 µM), and the optical incubation time of 120 min for color development, as shown in Figure S1 and the Supplementary Material.

ECAIA for Detecting DON
To achieve optimal performance of the ECAIA for DON detection, the optimal scFv-ALP concentration (12.5 µg/mL) (Table S2) and optimal coating DON-BSA concentration (1 µg/mL) (Figure 4A) were determined using checkerboard titration and direct competition assay, respectively.Based on these results, a series of conditions such as ionic strength, pH, MeOH content in the buffer, and competition time were optimized using IC 50 as the evaluation standard.As shown in Figure 4B, the 25 mM TBS buffer system provided the highest sensitivity (IC 50 = 134.45ng/mL).Similarly, the optimal reaction pH for the competition system was found to be 8.5 (IC 50 = 127.53ng/mL), 0% MeOH (IC 50 = 46.77ng/mL), and 45 min competition time (IC 50 = 8.43 ng/mL) (Figure 4C-E).
To achieve optimal performance of the ECAIA for DON detection, the optimal scFv-ALP concentration (12.5 µg/mL) (Table S2) and optimal coating DON-BSA concentration (1 µg/mL) (Figure 4A) were determined using checkerboard titration and direct competition assay, respectively.Based on these results, a series of conditions such as ionic strength, pH, MeOH content in the buffer, and competition time were optimized using IC50 as the evaluation standard.As shown in Figure 4B, the 25 mM TBS buffer system provided the highest sensitivity (IC50 = 134.45ng/mL).Similarly, the optimal reaction pH for the competition system was found to be 8.5 (IC50 = 127.53ng/mL), 0% MeOH (IC50 = 46.77ng/mL), and 45 min competition time (IC50 = 8.43 ng/mL) (Figure 4C-E).
Based on the optimized conditions, a standard curve of Log (DON concentration) vs. ΔB/B0 was plotted.The scatter plot was fitted with a logistic model, and the fitting equation was obtained as y = 0.11417 + [0.96659 − 0.11417]/[1 + (x/4.53891) 1.06584 ] (R 2 = 0.999).The linear range of ECAIA for DON detection was 1.2-35.41ng/mL (IC80-IC20), and the limit of detection (LOD) was 0.45 ng/mL (IC90), as shown in Figure 4F.Compared to the other reported immunoassays, ECAIA exhibited a wider linear range and higher sensitivity for DON detection (Table S3), indicating its broad potential in detecting other toxic small molecules in food.Based on the optimized conditions, a standard curve of Log (DON concentration) vs. ∆B/B 0 was plotted.The scatter plot was fitted with a logistic model, and the fitting equation was obtained as y = 0.11417 + [0.96659 − 0.11417]/[1 + (x/4.53891) 1.06584 ] (R 2 = 0.999).The linear range of ECAIA for DON detection was 1.2-35.41ng/mL (IC 80 -IC 20 ), and the limit of detection (LOD) was 0.45 ng/mL (IC 90 ), as shown in Figure 4F.Compared to the other reported immunoassays, ECAIA exhibited a wider linear range and higher sensitivity for DON detection (Table S3), indicating its broad potential in detecting other toxic small molecules in food.

Selectivity of the ECAIA
The present study investigated the selectivity of ECAIA towards four mycotoxins, namely OTA, AFB 1 , FB 1 , and ZEN.Four common mycotoxins, which were used as substitutes for DON, were diluted to various concentrations (0, 0.128, 0.64, 3.2, 16, 80, 400, and 2000 ng/mL) and subsequently tested using ECAIA.The results, as shown in Table 1, indicated that the IC 50 values for the four mycotoxins were >1200 ng/mL, and the cross- reactivity rates were <0.01%.Therefore, the ECAIA exhibits minimal cross-reactivity with those mycotoxins, suggesting good selectivity for the detection of DON.

Selectivity of the ECAIA
The present study investigated the selectivity of ECAIA towards four mycotoxins, namely OTA, AFB1, FB1, and ZEN.Four common mycotoxins, which were used as substitutes for DON, were diluted to various concentrations (0, 0.128, 0.64, 3.2, 16, 80, 400, and 2000 ng/mL) and subsequently tested using ECAIA.The results, as shown in Table 1, indicated that the IC50 values for the four mycotoxins were >1200 ng/mL, and the cross-reactivity rates were <0.01%.Therefore, the ECAIA exhibits minimal cross-reactivity with those mycotoxins, suggesting good selectivity for the detection of DON.

Sample Analysis and Validation
Due to the wide distribution of DON in crops, the complex composition of crops can interfere with the sensitivity and accuracy of detection methods.Therefore, eliminating matrix interference is crucial for the construction of ECAIA.In this study, corn, a common cereal, was selected for the methodological evaluation of ECAIA.The results showed that a 10-fold dilution of the corn extract could directly eliminate matrix interference (Figure S2).In addition, to further validate the effectiveness of ECAIA, DON was spiked into negative corn samples at different concentrations (200, 400, and 800 µg/kg).The intra-assay recoveries of ECAIA ranged from 88.9% to 118.3%, with relative standard deviations (RSD) ranging from 2.8% to 5.2%.Additionally, the inter-assay average recoveries of ECAIA ranged from 80.1% to 109.2%, with corresponding RSD ranging from 3.5% to 10.3% (Table 2).These results indicate that ECAIA has good accuracy when applied to the detection of DON in complex matrices.Furthermore, in this study, 10 positive samples were detected using both ECAIA and HPLC-UVD.As shown in Table S4, there was no significant The present study investigated the selectivity of ECAIA towards four mycotoxins, namely OTA, AFB1, FB1, and ZEN.Four common mycotoxins, which were used as substitutes for DON, were diluted to various concentrations (0, 0.128, 0.64, 3.2, 16, 80, 400, and 2000 ng/mL) and subsequently tested using ECAIA.The results, as shown in Table 1, indicated that the IC50 values for the four mycotoxins were >1200 ng/mL, and the cross-reactivity rates were <0.01%.Therefore, the ECAIA exhibits minimal cross-reactivity with those mycotoxins, suggesting good selectivity for the detection of DON.

Sample Analysis and Validation
Due to the wide distribution of DON in crops, the complex composition of crops can interfere with the sensitivity and accuracy of detection methods.Therefore, eliminating matrix interference is crucial for the construction of ECAIA.In this study, corn, a common cereal, was selected for the methodological evaluation of ECAIA.The results showed that a 10-fold dilution of the corn extract could directly eliminate matrix interference (Figure S2).In addition, to further validate the effectiveness of ECAIA, DON was spiked into negative corn samples at different concentrations (200, 400, and 800 µg/kg).The intra-assay recoveries of ECAIA ranged from 88.9% to 118.3%, with relative standard deviations (RSD) ranging from 2.8% to 5.2%.Additionally, the inter-assay average recoveries of ECAIA ranged from 80.1% to 109.2%, with corresponding RSD ranging from 3.5% to 10.3% (Table 2).These results indicate that ECAIA has good accuracy when applied to the detection of DON in complex matrices.Furthermore, in this study, 10 positive samples were detected using both ECAIA and HPLC-UVD.As shown in Table S4, there was no significant The present study investigated the selectivity of ECAIA towards four mycotoxins, namely OTA, AFB1, FB1, and ZEN.Four common mycotoxins, which were used as substitutes for DON, were diluted to various concentrations (0, 0.128, 0.64, 3.2, 16, 80, 400, and 2000 ng/mL) and subsequently tested using ECAIA.The results, as shown in Table 1, indicated that the IC50 values for the four mycotoxins were >1200 ng/mL, and the cross-reactivity rates were <0.01%.Therefore, the ECAIA exhibits minimal cross-reactivity with those mycotoxins, suggesting good selectivity for the detection of DON.

Sample Analysis and Validation
Due to the wide distribution of DON in crops, the complex composition of crops can interfere with the sensitivity and accuracy of detection methods.Therefore, eliminating matrix interference is crucial for the construction of ECAIA.In this study, corn, a common cereal, was selected for the methodological evaluation of ECAIA.The results showed that a 10-fold dilution of the corn extract could directly eliminate matrix interference (Figure S2).In addition, to further validate the effectiveness of ECAIA, DON was spiked into negative corn samples at different concentrations (200, 400, and 800 µg/kg).The intra-assay recoveries of ECAIA ranged from 88.9% to 118.3%, with relative standard deviations (RSD) ranging from 2.8% to 5.2%.Additionally, the inter-assay average recoveries of ECAIA ranged from 80.1% to 109.2%, with corresponding RSD ranging from 3.5% to 10.3% (Table 2).These results indicate that ECAIA has good accuracy when applied to the detection of DON in complex matrices.Furthermore, in this study, 10 positive samples were detected using both ECAIA and HPLC-UVD.As shown in Table S4, there was no significant The present study investigated the selectivity of ECAIA towards four mycotoxins, namely OTA, AFB1, FB1, and ZEN.Four common mycotoxins, which were used as substitutes for DON, were diluted to various concentrations (0, 0.128, 0.64, 3.2, 16, 80, 400, and 2000 ng/mL) and subsequently tested using ECAIA.The results, as shown in Table 1, indicated that the IC50 values for the four mycotoxins were >1200 ng/mL, and the cross-reactivity rates were <0.01%.Therefore, the ECAIA exhibits minimal cross-reactivity with those mycotoxins, suggesting good selectivity for the detection of DON.

Sample Analysis and Validation
Due to the wide distribution of DON in crops, the complex composition of crops can interfere with the sensitivity and accuracy of detection methods.Therefore, eliminating matrix interference is crucial for the construction of ECAIA.In this study, corn, a common cereal, was selected for the methodological evaluation of ECAIA.The results showed that a 10-fold dilution of the corn extract could directly eliminate matrix interference (Figure S2).In addition, to further validate the effectiveness of ECAIA, DON was spiked into negative corn samples at different concentrations (200, 400, and 800 µg/kg).The intra-assay recoveries of ECAIA ranged from 88.9% to 118.3%, with relative standard deviations (RSD) ranging from 2.8% to 5.2%.Additionally, the inter-assay average recoveries of ECAIA ranged from 80.1% to 109.2%, with corresponding RSD ranging from 3.5% to 10.3% (Table 2).These results indicate that ECAIA has good accuracy when applied to the detection of DON in complex matrices.Furthermore, in this study, 10 positive samples were detected using both ECAIA and HPLC-UVD.As shown in Table S4, there was no significant >1200 <0.01 ZEN namely OTA, AFB1, FB1, and ZEN.Four common mycotoxins, which were used as substitutes for DON, were diluted to various concentrations (0, 0.128, 0.64, 3.2, 16, 80, 400, and 2000 ng/mL) and subsequently tested using ECAIA.The results, as shown in Table 1, indicated that the IC50 values for the four mycotoxins were >1200 ng/mL, and the cross-reactivity rates were <0.01%.Therefore, the ECAIA exhibits minimal cross-reactivity with those mycotoxins, suggesting good selectivity for the detection of DON.

Sample Analysis and Validation
Due to the wide distribution of DON in crops, the complex composition of crops can interfere with the sensitivity and accuracy of detection methods.Therefore, eliminating matrix interference is crucial for the construction of ECAIA.In this study, corn, a common cereal, was selected for the methodological evaluation of ECAIA.The results showed that a 10-fold dilution of the corn extract could directly eliminate matrix interference (Figure S2).In addition, to further validate the effectiveness of ECAIA, DON was spiked into negative corn samples at different concentrations (200, 400, and 800 µg/kg).The intra-assay recoveries of ECAIA ranged from 88.9% to 118.3%, with relative standard deviations (RSD) ranging from 2.8% to 5.2%.Additionally, the inter-assay average recoveries of ECAIA ranged from 80.1% to 109.2%, with corresponding RSD ranging from 3.5% to 10.3% (Table 2).These results indicate that ECAIA has good accuracy when applied to the detection of DON in complex matrices.Furthermore, in this study, 10 positive samples were detected using both ECAIA and HPLC-UVD.As shown in Table S4, there was no significant >1200 <0.01

Sample Analysis and Validation
Due to the wide distribution of DON in crops, the complex composition of crops can interfere with the sensitivity and accuracy of detection methods.Therefore, eliminating matrix interference is crucial for the construction of ECAIA.In this study, corn, a common cereal, was selected for the methodological evaluation of ECAIA.The results showed that a 10-fold dilution of the corn extract could directly eliminate matrix interference (Figure S2).In addition, to further validate the effectiveness of ECAIA, DON was spiked into negative corn samples at different concentrations (200, 400, and 800 µg/kg).The intraassay recoveries of ECAIA ranged from 88.9% to 118.3%, with relative standard deviations (RSD) ranging from 2.8% to 5.2%.Additionally, the inter-assay average recoveries of ECAIA ranged from 80.1% to 109.2%, with corresponding RSD ranging from 3.5% to 10.3% (Table 2).These results indicate that ECAIA has good accuracy when applied to the detection of DON in complex matrices.Furthermore, in this study, 10 positive samples were detected using both ECAIA and HPLC-UVD.As shown in Table S4, there was no significant difference in the DON content of the same positive sample detected by ECAIA and HPLC-UVD.The detection limit for HPLC-UVD and ECAIA is 50 and 18 µg/kg, respectively, thus demonstrating that ECAIA exhibits higher sensitivity.Regression analysis of the DON content detected by ECAIA and HPLC-UVD yielded a linear equation of y = 0.94x + 14.85 (R 2 = 0.97) (Figure S3).In summary, the developed ECAIA has good sensitivity, accuracy, and stability, and can be used for DON detection in cereals.a Each assay was performed in triplicate on the same day.b Each assay was performed in triplicate for three days.

Conclusions
In this study, we report an enzyme cascade amplification immunoassay based on the scFv-ALP and the MnO 2 -TMB sensing system for the detection of DON in cereals.The bifunctional antibody scFv-ALP, which generates AA by hydrolysis of AAP, is used to link the antibody-antigen recognition system with the MnO 2 -TMB sensing system, thereby regulating the colorimetric system and achieving high-sensitivity detection of DON.By optimizing the relevant conditions for the MnO 2 -TMB sensing system and the competitive system, the LOD of DON in the ECAIA was achieved at 0.45 ng/mL.Given its high sensitivity and specificity in detecting DON, this immunoassay could become a standard tool for monitoring mycotoxins in cereals and other food products.Its adaptability as a detection model for other mycotoxins suggests that it could contribute significantly to reducing the risk of foodborne diseases caused by contaminated foods.With further optimization and validation, the ECAIA could become a routine analytical method in food testing laboratories worldwide.Furthermore, apart from MnO 2 nanosheets, other novel two-dimensional nanosheets such as MoS 2 , graphitic carbon nitride, layered double hydroxides, etc., can be introduced to enhance catalytic performance.From a methodological perspective, techniques like photonic immobilization [43] and surface-enhanced Raman spectroscopy can be combined with immunoassays to improve detection sensitivity and broaden their applicability.

Figure 1 .
Figure 1.Characterization of fusion protein scFv-ALP.(A) SDS-PAGE analysis of the purified scFv-ALP.Lane M: prestained protein marker; lane 1: purified scFv-ALP with a molecular weight of 81 kDa.(B) Analysis of the enzymatic activity and antibody functionality within scFv-ALP.Inset: diagrams of the reaction wells of the 96-well microplate.(C) Analysis of the detection performance of scFv-ALP by the direct competitive ELISA.Error bars denote the standard deviations (SDs) of three independent experiments.

Figure 1 .
Figure 1.Characterization of fusion protein scFv-ALP.(A) SDS-PAGE analysis of the purified scFv-ALP.Lane M: prestained protein marker; lane 1: purified scFv-ALP with a molecular weight of 81 kDa.(B) Analysis of the enzymatic activity and antibody functionality within scFv-ALP.Inset: diagrams of the reaction wells of the 96-well microplate.(C) Analysis of the detection performance of scFv-ALP by the direct competitive ELISA.Error bars denote the standard deviations (SDs) of three independent experiments.

Figure 4 .
Figure 4. Development of the ECAIA for DON.Optimization of DON-BSA concentration (A), ionic strength (B), pH (C), MeOH concentration (D), and competitive time (E) for ECAIA.(F) The standard competitive inhibition curve of ECAIA.The error bar represents the standard deviation from three independent experiments.

Figure 4 .
Figure 4. Development of the ECAIA for DON.Optimization of DON-BSA concentration (A), ionic strength (B), pH (C), MeOH concentration (D), and competitive time (E) for ECAIA.(F) The standard competitive inhibition curve of ECAIA.The error bar represents the standard deviation from three independent experiments.

Table 2 .
Recoveries of DON from the spiked cereals samples determined by ECAIA.