Fe3O4@Au Core–Shell Magnetic Nanoparticles for the Rapid Analysis of E. coli O157:H7 in an Electrochemical Immunoassay

Escherichia coli (E. coli) O157:H7 is a pathogenic bacterium that causes serious toxic effects in the human gastrointestinal tract. In this paper, a method for its effective analytical control in a milk sample was developed. To perform rapid (1 h) and accurate analysis, monodisperse Fe3O4@Au magnetic nanoparticles were synthesized and used in an electrochemical sandwich-type magnetic immunoassay. Screen-printed carbon electrodes (SPCE) were used as transducers, and electrochemical detection was performed by chronoamperometry using a secondary horseradish peroxidase-labeled antibody and 3,3′,5,5′-tetramethylbenzidine. This magnetic assay was used to determine the E. coli O157:H7 strain in the linear range from 20 to 2 × 106 CFU/mL, with a limit of detection of 20 CFU/mL. The selectivity of the assay was tested using Listeria monocytogenes p60 protein, and the applicability of the assay was assessed by analyzing a commercial milk sample, demonstrating the usefulness of the synthesized nanoparticles in the developed magnetic immunoassay.


Introduction
E. coli O157:H7 is a pathogenic bacteria strain that causes severe illness in humans through contaminated food or water. This bacterium is a Shiga-like toxin producer and causes severe foodborne diseases, especially in young children, the elderly, and patients with sensitive immune systems [1,2]. The World Health Organization (WHO) states that intoxication by E. coli O157:H7 could lead to serious conditions such as hemorrhagic diarrhea and acute kidney failure. The transmission of this strain mainly occurs through the consumption of undercooked or contaminated foods and liquids. For this reason, several countries have implemented guidelines concerning the identification of E. coli O157:H7 in food products to ensure public safety [3,4]. Therefore, monitoring E. coli O157:H7 is a challenge in the food safety field.
The rapid detection of E. coli O157:H7 at low concentrations is also a big concern in other fields, such as in environmental applications, medicine, and pharmacy [5,6]. Various conventional microbiological strategies are commonly applied for the analysis and detection of pathogenic bacteria, including Tissue Culture Plate (TCP), Plate Count Enumeration Method, Quartz Crystal Micro-balance resonators (QCM), and Polymerase Chain Reaction (LM p60 protein), expressed in E. coli, was obtained from Adipogen Life Sciences (Füllinsdorf, Switzerland).
The E. coli O157:H7 capture antibody, the E. coli O157:H7 detection antibody, and the secondary HRP-labeled antibody are abbreviated as Ab-C, Ab-D, and HRP-Ab, respectively.
The following solutions were used in this work: 10 mM MES buffer, pH 6 (B1), for activation of the carboxylic groups; 10 mM PBS containing 137 mM NaCl, pH 8.3 (B2), for preparing the blocking solution; and 10 mM PBS containing 137 mM NaCl, pH 7.4 (B3), to prepare working solutions and the antigen (E. coli O157:H7). For the washing steps, Tween ® 20 (0.01%), a surfactant used to prevent non-specific biomolecule binding and to reduce background signals, was included in the washing buffers (B1-T, B2-T, B3-T). All the reagents were prepared using Milli-Q ultrapure water (resistivity 18.2 MΩ·cm, at 25 • C) unless stated otherwise.
All the steps of the sandwich immunomagnetic assay were performed at room temperature, under continuous mixing (950 rpm) using a HulaMixer (Thermo Fisher Scientific, Oslo, Norway), and protected from light. The washing steps consisted of the addition of 100 µL of the desired buffer containing Tween-20, followed by additional stirring for 2 min. A DynaMagTM-2 magnetic rack (Life Technologies, Oslo, Norway) was used for the MNPs' attraction and separation of the supernatant in the various steps of the assay.
The chronoamperometric measurements were performed on SPCEs (DRP-110, Metrohm DropSens, Oviedo, Spain) that were connected (connector cable, DRP-CAC, Metrohm DropSens, Oviedo, Spain) to a potentiostat/galvanostat (PGSTAT 101, Metrohm Autolab, Utrecht, The Netherlands). A specific base containing a 4 mm diameter magnet was used to support the SPCE for precise magnetic attraction of the MNPs to the working electrode (WE) of the SPCE. All the measurements were performed in triplicate.

Preparation of E. coli O157:H7
E. coli O157:H7 was cultured in 50 mL of Tryptic Soy Broth (TSB) medium at 37 • C and an agitation speed of 240 rpm overnight. Next, the culture medium, including bacteria, was centrifuged at 1000 rpm for 5 min, and then the pellet was resuspended in 5 mL of PBS. To inactivate the bacteria for safe handling, the cells were heated at 100 • C for 15 min. Samples were frozen until use.

Fe 3 O 4 @Au Synthesis and Functionalization
The Fe 3 O 4 @Au MNPs were synthesized and functionalized following our previously reported method [19]. It consisted of (i) the synthesis of Fe 3 O 4 MNPs by thermal decomposition, followed by (ii) the coating of the Fe 3 O 4 MNPs with a gold shell by chemical reduction of Au(III) using oleylamine as reducing and capping agent, and (iii) the functionalization of the resulting Fe 3 O 4 @Au core-shell nanoparticles with a mixture of MCH and TOA (3:1 V/V). Briefly, NMP (35 mL), oleic acid (0.3 M), 1-hexadecanol (0.3 M), and oleylamine (0.3 M) were mixed and heated to 200 • C (under stirring and inert atmosphere). Then, [Fe(acac) 3 ] (0.15 M, 10 mL) was quickly added to this solution, and stirring was continued for 1 h. Then, the reaction mixture was cooled and kept under stirring overnight. The resulting MNPs were precipitated by adding ethanol (50 mL), and the material was washed several times with ethanol (to remove the excess of oleylamine and oleic acid) and redispersed in anhydrous toluene (5 mL (20 mL) and heated to 100 • C under an inert atmosphere. Subsequently, a solution containing HAuCl 4 .3H 2 O, oleylamine (7.07 mL), and anhydrous toluene (35 mL) was added dropwise under vigorous stirring to the pre-heated dispersion. The reaction mixture was stirred for 1 h. Then, the system was cooled to room temperature, and absolute ethanol (50 mL) was added. The coated nanoparticles were magnetically separated and washed several times with ethanol. The final nanomaterial was redispersed in anhydrous toluene (10 mL), and a dark red-purple dispersion was obtained.
The Fe 3 O 4 @Au concentration was determined and expressed in mg/mL. The functionalization of the Fe 3 O 4 @Au MNPs (4 mg/mL) was performed using ethanolic solutions of MCH (0.1 M) and TOA (0.1 M). The ethanolic thiol mixture with a volume ratio of 3:1 was added to the Fe 3 O 4 @Au dispersion and kept under stirring overnight. Then, the supernatant was removed (using a magnet), and the MNPs were dispersed in B1.

Magnetic Nanoparticles Characterization
The MNPs were characterized by transmission electron microscopy (TEM) on a JEOL-2000 Ex II TEM (Japan) equipment (JEOL, Tokyo, Japan). The procedure involved dispersing the samples in toluene under sonication and subsequently immersing a carbon-coated 200-mesh copper grid in the suspension, followed by air-drying. The average particle sizes and size distributions were calculated from the diameters of at least 100 particles randomly selected from the TEM micrographs. The dynamic light scattering (DLS) measurements were performed at 25 • C on a Zetasizer Nano ZS (Zetasizer, Malvern, UK).

Immobilization of Capture Antibody on Fe 3 O 4 @Au
The immobilization of the Ab-C on the Fe 3 O 4 @Au MNPs consisted of the following steps: 50 mg of Fe 3 O 4 @Au MNPs (0.25 mg/mL) functionalized with MCH-TOA were washed, followed by the separation of the modified Fe 3 O 4 @Au MNPs using the magnetic rack and disposal of the supernatant after 1 min. Then, the free carboxylic groups of the modified Fe 3 O 4 @Au MNPs were activated by mixing with 100 µL solution of EDC (200 mM) and NHS (50 mM) prepared in 500 µL of B1 buffer for 15 min under stirring. The supernatant was removed, and 100 µL of Ab-C (10 µg/mL, in B1) were added for 1 h, followed by a washing step with B1-T and B2-T. Finally, 100 µL of EA (0.1 M, in B2) were added and incubated for 10 min to block the unreacted carboxylic groups. Subsequent washing steps with B2-T and B3-T were carried out, and the Ab-C-modified Fe 3 O 4 @Au MNPs were finally dispersed in B3-T and stored at 4 • C until use. Figure 1 illustrates the different steps of the biofunctionalization, optimized immunoassay, and electrochemical detection strategy. In the optimized assay, a 10-µL aliquot of the Ab-C-modified Fe 3 O 4 @Au MNPs dispersion was taken, the supernatant was removed, and 100 µL of an E. coli O157:H7 standard/sample solution was added and incubated for 30 min. Then, 100 µL of a solution containing both Ab-D (5.0 µg/mL) and HRP-Ab (initial concentration: 1 g/L; 2000× dilution in B3) were added and incubated for 30 min. A final washing step was performed with B3-T, and the suspension was maintained in B3-T. This suspension (10 µL) was placed on the SPCE, the MNPs were attracted to the working electrode (WE) with a magnet, and the supernatant was removed. The biorecognition event was monitored by adding the enzymatic substrate TMB (40 µL, 1 min, and absence of light) and subsequently recording a chronoamperogram (at 0 V for 1 min). During the enzymatic reaction, HRP catalyzes the oxidation of TMB, which is then detected by chronoamperometry [23]. min. A final washing step was performed with B3-T, and the suspension was maintained in B3-T. This suspension (10 µL) was placed on the SPCE, the MNPs were attracted to the working electrode (WE) with a magnet, and the supernatant was removed. The biorecognition event was monitored by adding the enzymatic substrate TMB (40 µL, 1 min, and absence of light) and subsequently recording a chronoamperogram (at 0 V for 1 min). During the enzymatic reaction, HRP catalyzes the oxidation of TMB, which is then detected by chronoamperometry [23].

Milk Sample Analysis
To evaluate the applicability of the proposed assay, a milk sample was purchased from a local supermarket (Central Lechera Asturiana, Asturias, Spain) and used without any pretreatment. The milk sample was spiked with different E. coli O157:H7 concentrations (2 × 10 3 , 2 × 10 2 , and 2 × 10 1 CFU/mL), diluted (1:10), and analyzed. Control samples (non-spiked milk) were also included in the analysis.

Fe3O4@Au MNP Synthesis and Characterization
The magnetic Fe3O4 cores were synthesized by the thermal decomposition method [19]. To prevent the oxidation of the Fe3O4 MNPs and to use them as platform for the development of the magnetic immunoassay, they were coated with a gold shell. The coating process involved a gradual deposition of gold onto the surface of the Fe3O4 MNPs by a controlled temperature-induced reduction of Au(III) to Au(0) using oleylamine. In this process, oleylamine functions both as a reducing agent and as a stabilizer of the resulting core-shell MNPs, preventing the agglomeration of the nanoparticles and maintaining their stability [24]. Figure 2 shows the TEM micrographs of Fe3O4 and Fe3O4@Au MNPs. The TEM images revealed that the Fe3O4 MNPs were well dispersed, presenting a lognormal particle size distribution and an average particle size of 5.4 nm (σ = 1.0). The

Milk Sample Analysis
To evaluate the applicability of the proposed assay, a milk sample was purchased from a local supermarket (Central Lechera Asturiana, Asturias, Spain) and used without any pretreatment. The milk sample was spiked with different E. coli O157:H7 concentrations (2 × 10 3 , 2 × 10 2 , and 2 × 10 1 CFU/mL), diluted (1:10), and analyzed. Control samples (non-spiked milk) were also included in the analysis.

Fe 3 O 4 @Au MNP Synthesis and Characterization
The magnetic Fe 3 O 4 cores were synthesized by the thermal decomposition method [19]. To prevent the oxidation of the Fe 3 O 4 MNPs and to use them as platform for the development of the magnetic immunoassay, they were coated with a gold shell. The coating process involved a gradual deposition of gold onto the surface of the Fe 3 O 4 MNPs by a controlled temperature-induced reduction of Au(III) to Au(0) using oleylamine. In this process, oleylamine functions both as a reducing agent and as a stabilizer of the resulting core-shell MNPs, preventing the agglomeration of the nanoparticles and maintaining their stability [24]. Figure 2 shows the TEM micrographs of Fe 3 O 4 and Fe 3 O 4 @Au MNPs. The TEM images revealed that the Fe 3 O 4 MNPs were well dispersed, presenting a log-normal particle size distribution and an average particle size of 5.4 nm (σ = 1.0). The Fe 3 O 4 @Au MNPs had a nearly spherical shape and a particle size of 13.7 ± 1.9 nm, confirming the coating of the Fe 3 O 4 cores with gold (Table 1) (Table 1). In both cases, the solvodynamic particle sizes are consistent with the corresponding particle sizes obtained by TEM, indicating the lack of agglomeration of the particles in the solution.
urements were carried out to confirm the coating of the Fe3O4 MNPs with gold and obtain information on the colloidal stability of the nanomaterials (Figure 3). The Fe3O4 MNPs dispersed in toluene presented an average solvodynamic diameter of 5.6 nm, while the Fe3O4@Au MNPs showed an average solvodynamic diameter of 15.7 nm (Table 1). In both cases, the solvodynamic particle sizes are consistent with the corresponding particle sizes obtained by TEM, indicating the lack of agglomeration of the particles in the solution.

Optimization of Experimental Conditions
For the development of a successful immunoassay, the optimization of the experimental parameters is fundamental [25,26]. Therefore, the variables of interest, such as the working solution, incubation time, and reagents' concentrations, were optimized according to their relevance to the immunoassay. The selected values were chosen according to the signal-to-blank ratio (S/B) using the current intensity values measured in the absence and in the presence of 2 × 10 6 CFU/mL of E. coli O157:H7. The obtained data correspond to the average and standard deviation of three replicates.
To obtain selective detection of the target bacteria, it is necessary to minimize the background signal related to the non-specific reaction of biomolecules on the sensor platform. Therefore, the effect of adding BSA to the working solutions was evaluated. B3-T and B3-T-BSA 1% (m/V) were tested and compared using the following conditions: Ab-Cmodified Fe3O4@Au MNPs volume: 10 µL; Ab-C concentration: 10 µg/mL; E. coli O157:H7 concentration: 2×10 6 CFU/mL; Ab-D concentration: 5 µg/mL, HRP-Ab: 2000× dilution; and incubation time: 60 min. As can be observed in Figure 4A, the S/B ratio improved when the immune reactions in different steps were performed in the working solution containing BSA 1% (m/V). The addition of BSA clearly reduced the blank signal more than the signal obtained in the presence of bacteria, improving the S/B ratio. Therefore, B3-T-BSA 1% (m/V) was selected to proceed with the studies.
To improve the total assay time, the 'step-by-step assay' approach was tested using different incubation times (30 or 60 min): incubation of (1) bacteria, 30 min; Ab-D, 30 min; and HRP-Ab 30 min; (2) bacteria, 60 min; Ab-D, 60 min, and HRP-Ab, 30 min. In Figure  4B, it is possible to observe that the increase in the incubation time (assay 2) decreased the S/B ratio. Therefore, an incubation time of 30 min for each biomolecule (bacteria, Ab-D, and HRP-Ab) was used in the subsequent studies.
The Ab-D concentration (2.5, 5.0, and 15 µg/mL) was optimized under the previously mentioned conditions. Figure 4C shows that when the Ab-D concentration increased from 2.5 to 5.0 µg/mL, the S/B ratio also increased; however, for the highest concentration (15 µg/mL), the S/B ratio decreased, which indicates that saturation was attained. Therefore, 5.0 µg/mL of Ab-D was chosen to proceed with the development of the assay. The dilution of the HRP-Ab (1000×, 2000×, and 4000×) was optimized under the previously optimized conditions. The obtained results ( Figure 4D) show that higher dilutions led to an increase in the S/B ratio but with lower bacteria and blank signals. Therefore, the intermediate dilution of 2000× was considered adequate.

Optimization of Experimental Conditions
For the development of a successful immunoassay, the optimization of the experimental parameters is fundamental [25,26]. Therefore, the variables of interest, such as the working solution, incubation time, and reagents' concentrations, were optimized according to their relevance to the immunoassay. The selected values were chosen according to the signal-to-blank ratio (S/B) using the current intensity values measured in the absence and in the presence of 2 × 10 6 CFU/mL of E. coli O157:H7. The obtained data correspond to the average and standard deviation of three replicates.
To obtain selective detection of the target bacteria, it is necessary to minimize the background signal related to the non-specific reaction of biomolecules on the sensor platform. Therefore, the effect of adding BSA to the working solutions was evaluated. B3-T and B3-T-BSA 1% (m/V) were tested and compared using the following conditions: Ab-C-modified Fe 3 O 4 @Au MNPs volume: 10 µL; Ab-C concentration: 10 µg/mL; E. coli O157:H7 concentration: 2 × 10 6 CFU/mL; Ab-D concentration: 5 µg/mL, HRP-Ab: 2000× dilution; and incubation time: 60 min. As can be observed in Figure 4A, the S/B ratio improved when the immune reactions in different steps were performed in the working solution containing BSA 1% (m/V). The addition of BSA clearly reduced the blank signal more than the signal obtained in the presence of bacteria, improving the S/B ratio. Therefore, B3-T-BSA 1% (m/V) was selected to proceed with the studies.
To improve the total assay time, the 'step-by-step assay' approach was tested using different incubation times (30 or 60 min): incubation of (1) bacteria, 30 min; Ab-D, 30 min; and HRP-Ab 30 min; (2) bacteria, 60 min; Ab-D, 60 min, and HRP-Ab, 30 min. In Figure 4B, it is possible to observe that the increase in the incubation time (assay 2) decreased the S/B ratio. Therefore, an incubation time of 30 min for each biomolecule (bacteria, Ab-D, and HRP-Ab) was used in the subsequent studies.
The Ab-D concentration (2.5, 5.0, and 15 µg/mL) was optimized under the previously mentioned conditions. Figure 4C shows that when the Ab-D concentration increased from 2.5 to 5.0 µg/mL, the S/B ratio also increased; however, for the highest concentration (15 µg/mL), the S/B ratio decreased, which indicates that saturation was attained. Therefore, 5.0 µg/mL of Ab-D was chosen to proceed with the development of the assay. The dilution of the HRP-Ab (1000×, 2000×, and 4000×) was optimized under the previously optimized conditions. The obtained results ( Figure 4D) show that higher dilutions led to an increase in the S/B ratio but with lower bacteria and blank signals. Therefore, the intermediate dilution of 2000× was considered adequate.
Regarding the optimization of the total assay time, a combined (joined) step protocol, comprising the preincubation of Ab-D (5.0 µg/mL) and HRP-Ab (2000×) in a single solution (15 min before use in the assay), was compared with the step-by-step procedure. The obtained data ( Figure 4E) shows that the joined step strategy led to the highest S/B ratio. It also decreased the assay time from 1.5 h to 1 h and the number of incubation steps, so this procedure was adopted for the following studies. Regarding the optimization of the total assay time, a combined (joined) step protocol, comprising the preincubation of Ab-D (5.0 µg/mL) and HRP-Ab (2000×) in a single solution (15 min before use in the assay), was compared with the step-by-step procedure. The obtained data ( Figure 4E) shows that the joined step strategy led to the highest S/B ratio. It also decreased the assay time from 1.5 h to 1 h and the number of incubation steps, so this procedure was adopted for the following studies.

Analytical Performance of the Immunoassay for E. coli O157:H7 Analysis
To test the performance of the developed immunoassay, different solutions of E. coli O157:H7 were prepared (2 × 10 1 , 2 × 10 2 , 2 × 10 3 , 2 × 10 4 , 2 × 10 5 , 2 × 10 6 , and 2 × 10 7 CFU/mL) and analyzed using the optimized conditions. The calibration straight was established between 2 × 10 1 and 2 × 10 6 CFU/mL, with the following linear regression equation: i (µA) = 7.1 × 10 −7 [E. coli] (CFU/mL) + 0.21, r 2 = 0.998, n = 6 ( Figure 5A). The limit of detection (LOD = 20 CFU/mL) was calculated according to the equation 3.3 × σ/S, where σ is the standard deviation of the intercept, and S is the slope of the calibration line. Furthermore, the coefficient of variation of the method (9.9%) demonstrated the good precision of the method. The analytical characteristics allow us to conclude that the developed immunoassay could be applied to the analysis of E. coli in food samples. tween 2 × 10 and 2 × 10 CFU/mL, with the following linear regression equation: i (µA) = 7.1 × 10 −7 [E. coli] (CFU/mL) + 0.21, r 2 = 0.998, n = 6 ( Figure 5A). The limit of detection (LOD = 20 CFU/mL) was calculated according to the equation 3.3 × σ/S, where σ is the standard deviation of the intercept, and S is the slope of the calibration line. Furthermore, the coefficient of variation of the method (9.9%) demonstrated the good precision of the method. The analytical characteristics allow us to conclude that the developed immunoassay could be applied to the analysis of E. coli in food samples.
Furthermore, the selectivity of the immunoassay was tested against a cell surface protein (p60) of Listeria monocytogenes (1000 ng/mL) that is secreted in large quantities into growth media. As shown in Figure 5B, the current intensity obtained for E. Coli O157:H7 was significantly higher than the one obtained for the Listeria monocytogenes protein, thus demonstrating that good selectivity was achieved. According to the obtained results, an additional washing step or the adjustment of the Tween-20 amount in the washing buffer could allow a more effective removal of the non-target protein, reducing the corresponding signal. However, this would lead to a more laborious procedure.

Milk Sample Analysis
To assess the accuracy of the assay's results, liquid milk was selected to evaluate the application of the proposed assay for rapid detection (1 h assay) of E. coli O157:H7 in foods. The sample was prepared as explained in Section 2.7, "milk sample analysis", and spiked with different concentrations of E. coli O157:H7 (0; 2 × 10 1 , 2 × 10 2 , and 2 × 10 3 CFU/mL). E. coli O157:H7 was only quantifiable in the 2 × 10 2 and 2 × 10 3 CFU/mL solutions. Acceptable recoveries for these concentrations were obtained (110% and 125%) with adequate precision (the coefficient of variation of the results (CV) was lower than 3%) (Table 2). Although a LOD of 20 CFU/mL was achieved in buffer solution, bacteria analysis in food samples (such as milk) is always a challenge because of the complexity of the sample matrix. Therefore, for lower bacteria concentrations, a lower analytical signal than the one obtained for the LOD in buffer solution was obtained. Furthermore, the selectivity of the immunoassay was tested against a cell surface protein (p60) of Listeria monocytogenes (1000 ng/mL) that is secreted in large quantities into growth media. As shown in Figure 5B, the current intensity obtained for E. Coli O157:H7 was significantly higher than the one obtained for the Listeria monocytogenes protein, thus demonstrating that good selectivity was achieved. According to the obtained results, an additional washing step or the adjustment of the Tween-20 amount in the washing buffer could allow a more effective removal of the non-target protein, reducing the corresponding signal. However, this would lead to a more laborious procedure.

Milk Sample Analysis
To assess the accuracy of the assay's results, liquid milk was selected to evaluate the application of the proposed assay for rapid detection (1 h assay) of E. coli O157:H7 in foods. The sample was prepared as explained in Section 2.7, "milk sample analysis", and spiked with different concentrations of E. coli O157:H7 (0; 2 × 10 1 , 2 × 10 2 , and 2 × 10 3 CFU/mL). E. coli O157:H7 was only quantifiable in the 2 × 10 2 and 2 × 10 3 CFU/mL solutions. Acceptable recoveries for these concentrations were obtained (110% and 125%) with adequate precision (the coefficient of variation of the results (CV) was lower than 3%) ( Table 2). Although a LOD of 20 CFU/mL was achieved in buffer solution, bacteria analysis in food samples (such as milk) is always a challenge because of the complexity of the sample matrix. Therefore, for lower bacteria concentrations, a lower analytical signal than the one obtained for the LOD in buffer solution was obtained. 3.5. Comparison with Other Electrochemical Immunoassays for the Determination of E. coli Table 3 shows a comparison between the magnetic immunoassay developed in this work and reported electrochemical sensors/assays for E. coli analysis. The Fe 3 O 4 @Au MNPs have several advantages due to their unique combination of SPIONs and a gold shell. Firstly, the magnetic cores allow the efficient collection of the analyte during the assay, which enhances the sensitivity of the assay and reduces the signal background signal because of the elimination of possible interferents during the washing step. Secondly, the gold shell of the Fe 3 O 4 @Au MNPs provides several advantages, such as increased biocompatibility, chemical stability, and ease of functionalization. The gold shell makes the system more versatile for functionalization and bioconjugation with different bioreceptors, such as antibodies or enzymes, allowing for a wide range of applications. The gold shell also enhances the stability of the MNPs in biological fluids and reduces the potential for aggregation. Compared to other electrochemical immunoassays reported in the literature, the developed magnetic immunoassay is cost-effective, portable, and user-friendly, without requiring expensive equipment. Additionally, the assay developed in this work demonstrated an acceptable LOD for E. coli O157:H7 within a short assay time, which is comparable with the works reported earlier. The benefits of this magnetic immunoassay include the elimination of complex immobilization procedures, ease of use, and analysis in a cost-effective assay, making it practical for the detection of E. coli O157:H7. Furthermore, this approach can be adapted for the detection of other pathogens by utilizing different bacterium-specific bio-elements.

Conclusions
In this work, an electrochemical magnetic immunoassay for the rapid analysis of E. coli O157:H7 bacteria was developed by using core-shell Fe 3 O 4 @Au MNPs. The coreshell MNPs were successfully synthesized by coating magnetic Fe 3 O 4 cores with a gold shell through the reduction of Au(III) into Au(0) on the surface of the cores promoted by oleylamine. The resulting Fe 3 O 4 @Au MNPs were functionalized with a mixture of 6-mercapto-1-hexanol and thioctic acid in ethanol. These magnetic platforms were used in a sandwich-type assay and were easily held on the surface of an SPCE with an external magnet.
The specific capture antibody was covalently bioconjugated to the carboxylated selfassembled monolayer on the surface of Fe 3 O 4 @Au MNPs, while a detection antibody and a secondary HRP-labeled antibody were also used in the assay. Chronoamperometric measurements allowed efficient bacteria analysis in the linear range between 20 and 2 × 10 6 CFU/mL. Recovery analysis demonstrated that at least 2 × 10 2 CFU/mL of E. coli could be efficiently detected in a milk sample. The developed immunoassay offers a costeffective, portable, and user-friendly solution, without the need for expensive equipment. It allows the detection of E. coli O157:H7 in a short time (1 h) with an acceptable LOD, which is a better performance when compared to other reported electrochemical assays. The immunoassay eliminates complex immobilization procedures, making it practical for the food industry to perform fast and reliable analysis. Moreover, it can be adapted for the detection of other pathogens by utilizing different bacterium-specific bio-elements.