Ratiometric electrochemical aptasensor based on split aptamer and Au-RGO for detection of Aflatoxin M1

A novel ratiometric electrochemical aptasensor based on split aptamer and Au-rGO nanomaterials was proposed to detect Aflatoxin M1 (AFM1). In this work, Au-rGO nanomaterials were coated on the electrode through the electrodeposition method to increase the aptamer enrichment. We split the aptamer of AFM1 into 2 sequences (S1 and S2), where sequence S1 was immobilized on the electrode due to the Au-S bond, whereas sequence S2 was tagged with methylene blue (MB) and acted as a response signal. A complementary strand to S1 (CS1) labeled with ferrocene (Fc) was introduced as another reporter. In the presence of AFM1, CS1 was released from the electrode surface due to the formation of the S1-AFM1-S2 complex, leading to a decrease in Fc and an increase in MB signal. The developed ratiometric aptasensor exhibited a linear range of 0.03 μg L −1 to 2.00 μg L −1 , with a detection limit of 0.015 μg L −1 for AFM1 detection. The ratiometric ap-tasensor also showed a linear relationship from 0.2 μg L −1 to 1.00 μg L −1 , with a detection limit of 0.05 μg L −1 in natural milk after sample pretreatment, indicating the successful application of the developed ratiometric aptasensor. Our proposed strategy provides a new way to construct aptasensors with high sensitivity and selectivity.


INTRODUCTION
Aflatoxin M1 (AFM1) contamination in milk and dairy products is a significant global issue.AFM1 was detected in a significant proportion of milk samples, including those undergoing pasteurization and ultrahigh temperature (UHT) treatment (Womack et al., 2016;Min et al., 2020;Topi et al., 2022).The presence of AFM1 poses a severe threat to human health (Frazzoli et al., 2017).The prolonged intake of AFM1 would lead to chronic hepatitis or even liver tumors.Thus, AFM1 has been classified as a group 1 carcinogen by the International Agency for Research on Cancer (IARC; Sharma et al., 2016;Liu et al., 2020).To avoid serious health hazards, the European Commission set a maximum level of 0.25 μg L −1 for AFM1 in milk, and the US Food and Drug Administration has established a limit of 0.5 μg L −1 for AFM1 in milk.However, AFM1 contamination continues to spread worldwide (Mousavi Khaneghah et al., 2021;Zebib et al., 2022).Therefore, developing effective and sensitive methods to detect AFM1 in dairy samples is of great significance.
Traditional rapid detection methods for AFM1, such as immunoaffinity chromatography (IAC) and enzymelinked immunosorbent assay (ELISA), have limitations in terms of cost, preparation time, and chemical stability (Bacher et al., 2012;Sun et al., 2022).Aptamers, short nucleic acid sequences selected in vitro through systematic evolution (SELEX), present a promising alternative for detecting AFM1 due to their high stability, specificity, and low cost.Recently, numerous aptasensor exhibiting advantages of high sensitivity and selectivity have been developed.Zhang et al. (2021) developed an aptasensor for AFM1 based on the colorimetric response of gold nanoparticles (AuNPs).An ultra-sensitive aptasensor for AFM1 was proposed (Zeng et al., 2018).Chloramphenicol was introduced

Ratiometric electrochemical aptasensor based on split aptamer and Au-RGO for detection of Aflatoxin M1
as an electrochemical reaction promoter to amplify the signal.The aptasensor exhibited a detection limit for AFM1 as low as 0.09 pg mL −1 with a linear range of 0.4 pg mL −1 to 400 ng mL −1 , demonstrating excellent sensitivity and stability.However, the flexible conformation of the aptamer may lead to a false positive.It is necessary to propose a viable solution for improving the performance of aptasensors, such as constructing a tetrahedral structure or splitting aptamer (R. O'Steen and M.Kolpashchikov, 2022;Shang et al., 2023).Wang et al. (2019)  The electrochemical aptasensors have drawn attention owing to their fast response, high sensitivity, and facile operation (Sadeghi et al., 2018).To further improve the sensitivity, nanomaterials were introduced (Bocanegra-Rodríguez et al., 2021;Alwarappan et al., 2022).Compared with individual nanomaterials, composite nanomaterials showed superior performance (Sharma et al., 2018). Samadi Pakchin et al. (2020) reported an electrochemical biosensor based on polyamidoamine/gold nanoparticles and 3-dimensional reduced graphene oxide-multiwall carbon nanotubes (3D rGO-MWCNTs) for the detection of cancer antigen 125.This biosensor used 3D rGO-MWCNTs to modify the glassy carbon electrode, enhancing its conductivity and specific surface area.It had a wide linear range (0.0005-75 U mL −1 ), and the detection limit was 6 mu U mL −1 .
Conventional electrochemical biosensors are based on the decrease or increase of a single signal.Due to the interference of changes in the microenvironmental, operational complexity, and other factors, it is limited in practical application.To overcome the deficiency of the single-signal readout, the ratiometric strategy of dual-signal output has been developed and employed in various fields, such as medicine and analysis.Based on the above considerations, we demonstrated a ratiometric electrochemical aptasensor based on split aptamer and composite nanomaterials for detecting AFM1.To reduce the false positive caused by the flexibility of the full AFM1 aptamer, the full aptamer was split into 2 sequences.The introduction of Au-rGO composites improved electroconductibility and increased surface area for enriching more split aptamer.The ratiometric electrochemical aptasensor was easily fabricated and exhibited high selectivity and sensitivity.
All electrochemical measurements were conducted on the CHI600E workstation (Shanghai CH Instruments, China).The 3-electrode cell was used, where the glassy carbon electrode (GCE) was chosen as the working electrode, the saturated calomel electrode as the reference electrode, and the platinum wire as the counter electrode.Scanning electron microscopy (SEM) was obtained from the Gemini SEM 300 scanning electron microscope (Carl Zeiss AG, Germany).Deionized water was purified using Millipore Milli-Q Ultrapure Water System (Bedford, MA, USA) and used for all experiments.

Pretreatment of Electrode
The 2 mm diameter glassy carbon electrode was polished with 0.3 μm and 0.05 μm alumina particles in water slurry for 10 min.Subsequently, the GCE was ultrasonically cleaned with ethanol and ultrapure water and dried using N 2 .

Preparation of Au-rGO Composites
RGO was dispersed in ultra-pure water and sonicated for 1 h to obtain a homogeneous rGO solution.Different concentrations of rGO solution were dropped on the electrode and then dried at room temperature.The pretreated GCE was used as the working electrode and immersed in the chloroauric acid solution (0.1 M KCl).The electrodeposition was carried out with constant voltage mode.The cyclic voltammetry (CV) curve of the electrode modified with Au-rGO composite nanomaterial was scanned in 0.5 M H 2 SO 4 solution, and the scanning range was −0.4 V -+1.6 V, and the scanning speed was 100 mV s −1 .The active area on the electrode surface was calculated using the reduction peak of AuO.The electrode was characterized by CV and electrochemical impedance (EIS) in the 0.1 M KCl solution with 5.0 mM potassium ferricyanide ([Fe (CN) 6 ] 3-/4-).

Fabrication of the ratiometric aptasensor
Ten μM SH-S1 was incubated with 1 mM TCEP for 1h to cleave the disulfide bonds.A concentration of 1 μM S1 was obtained by dilution with 200 μL Tris-HCl (pH 7.4).The prepared electrode was incubated with SH-S1 for 16 h to immobilize the S1 on the electrode by Au-S bonds.Subsequently, the electrode was washed with Tris-HCl (pH 7.4, 100 mM NaCl) to remove unbounded S1.The electrode was immersed in 1 mM MCH solution to block the uncovered sites.Then, the modified electrode was rinsed with Tris-HCl.Finally, the electrode was immersed in an 1 μM Fc-CS1 solution for a while.The aptasensor platform was established.The CV and EIS were measured to monitor each immobilization step.
Concentration of S1 was investigated.In the experiment, different concentrations of S1 (0.2 μM, 0.4 μM, 0.6 μM, 0.8 μM, 1.0 μM and 1.2 μM) was modified on the surface of the work electrode for the platform.

Analysis of Real Samples
The ratiometric electrochemical aptasensor was used to detect AFM1 in the skim milk and pure milk samples to demonstrate its practicability and accuracy.Ten mL milk was diluted with 85 mL Tris-HCl (pH = 7.4), mixed with 5 mL 18 mM calcium chloride solution, and centrifuged for 10 min at 6000 rpm.Then the supernatant was collected and detected by the developed aptasensor.

Principle of the ratiometric electrochemical aptasensor
In this work, the ratiometric electrochemical aptasensor based on split aptamer and Au-RGO nanocomposites was developed to detect the AFM1.The principle was illustrated in Scheme 1.The electrode was initially modified by the rGO using the dropping method, followed by immobilization of gold nanoparticles through electro-deposition.Additionally, based on our research, the AFM1 aptamer was split into 2 segments (S1 and S2) in a 1:1 ratio (Nguyen et al., 2013;Zhu et al., 2015;Guo et al., 2018).S1 was labeled with sulfate group (SH) and immobilized on the electrode via Au-S bond.S2 was modified methylene blue (MB) to serve as a sensing signal.Also, the complementary strand of S1 (CS1), tagged with ferrocene (Fc), was introduced as another sensing signal.CS1 was not involved in target recognition but competed with S2 + AFM1 for binding to S1.The hybridization between S1 and CS1 led to the proximity of Fc to the electrode and generated an Fc signal.In the presence of AFM1, CS1 was released from the electrode due to the formation of the S1-S2-AFM1 complex, resulting in an increased MB signal and a decreased Fc signal.In the absence of AFM1, a high peak current of Fc was observed due to double-stranded hybridization between S1 and CS1, leading to probe Fc close to the electrode.The MB signal was not measured because S2 was in the medium.In contrast, when AFM1 was present, the Fc peak decreased while the MB peak appeared.Furthermore, with the increase of AFM1 concentration, the Fc response signal gradually decreased, and the corresponding MB signal gradually increased.

Li et al.: Ratiometric electrochemical aptasensor…
Optimization of rGO.RGO was deposited on the electrode to increase the electrode surface area.As shown in Figure 1A, the peak current initially increased with the increasing concentration of rGO solution, illustrating that the electron transfer rate on the electrode surface was enhanced due to the coating of rGO (Sun et al., 2021).The peak current reached equilibrium with 3 mg mL −1 rGO, indicating that the glassy carbon electrode-loaded rGO had reached saturation.Therefore, 3 mg mL −1 rGO was chosen for further experiments.The prominent wrinkles observed in Figure 1B indicated the presence of rGO film coating on the electrode.

Fabrication of Au-rGO composite nanomaterial.
Deposition potential is a critical determinant for gold deposition.to the size of around 4 μm, attributing to the unstable energy of AuNPs (Lin et al., 2010).Under a negative deposition potential (D-E), the amount of AuNPs deposited on the rGO surface gradually decreased with the decrease of deposition potential.At −0.5 V, numerous bubbles appeared on the electrode, blocking the growth of AuNPs.The energy spectrum of the composite nanomaterial depicted in Figure 2F validated that the successful modification of Au-rGO nanomaterial on the electrode surface was achieved by confirming the presence of both Au and C elements.
The electrochemical active surface areas of the electrode (ESA) were also investigated by calculating the Coulomb peak areas of the gold reduction at around 0.9 V in 0.05 M H 2 SO 4 .As depicted in Figure 2G, multiple overlapping oxidation peaks appear at about 1.3 V and a single sharp reduction peak at about 0.9 V.This is a typical feature of undefiled gold surfaces (Liu et al., 2021;Shah et al., 2022;Yeh et al., 2022).Meanwhile, the active surface area increased rapidly with increasing deposition potential.A higher active area was observed at +0.3 V due to the conducive reduction of Au 2+ by positive potential.Therefore, the optimal deposition potential was +0.3 V.
The deposition time was optimized to obtain a stable and sizeable active surface area for aptamer immobilization.As shown in Figure 3A-D, the accumulation of AuNPs on the surface of rGO gradually increased, and the aggregation size correspondingly enlarged with prolonged deposition times.AuNPs tend to aggregate into clusters under a deposition time of 10 min (Figure 3A).Extending to 30 min, the irregularly shaped AuNPs underwent stacking and formed particles of 0.5 μm size (Figure 3B).When the deposition time was 60 min, the agglomeration AuNPs showed spherical shapes with an increased size of 1 μm (Figure 3C).Further extending the deposition time to 90 min, the spherical morphology was gradually destroyed, and transformed into the cauliflower-like structure with a size of 2 μm (Figure 3D).The growth pattern followed the continuous nucleation mechanism (Zhang et al., 2019).
The ESA obtained under different deposition times was calculated.It can be seen from Figure 3D that the ESA first increased and then slightly decreased with the increase of the deposition time.When the deposition time was 60 min, the maximum ESA was observed, attributing to the formation of homogeneous spheroidal AuNPs.Extending the time to 90 min resulted in a slight decrease in the ESA, attributed to the gradual transformation of spherical AuNPs into clustered ones, which subsequently covered active sites.Because the AuNPs with spherical structures gradually turned into AuNPs with clusters, and later-grown AuNPs covered the active sites (Qu et al., 2018;Shi and Wang, 2020;Yuan et al., 2023).Therefore, the deposition time of 60 min was chosen for further experiment.
The concentration of AuHCl 4 was investigated using morphology and ESA through the preparation of the composite nanomaterial under a potential of +0.3 V and deposition time of 60 min, as shown in Figure 4.The size of agglomeration AuNPs decreased from 3 μm to 1 μm as the concentration of AuHCl 4 increased from 2 mM to 10 mM (Figure 4A-C).When the concentration was 15 mM (Figure 4D), the new crystal nuclei were generated on the original nanoparticle structure (Ko et al., 2015), causing a slight increase in the size and destruction of the spherical structure.The results of ESA (Figure 4E) showed that the active surface area of the electrode also increased first and then decreased.The electrode with the largest active area was obtained using 10 mM AuHCl 4 , attributing to the relatively uniform spherical morphology of AuNPs.Therefore, 10 mM AuHCl 4 was the optimal concentration.
Electrochemical characterization of the aptasensor.The CV and EIS were used to monitor the aptasensor fabrication.Au-rGO has the potential to enhance material conductivity due to its high electronic transfer capability and large surface area (Zhang et al., 2019).As shown in Figure S1A, compared with the bare electrode (curve a), the peak current increased after Au-rGO (curve b) modification, indicating enhanced charge transfer.After immobilizing S1, the peak current decreased, and the peak width increased (curve c), demonstrating the blocking of electron transfer due to the coating of S1.Subsequently, the uncovered sites blocked with MCH further decreased the peak current (curve d).Curve e showed the peak current further decreased, demonstrating the hybridization between S1 and CS1.In the presence of AFM1 and S2, CS1 was released, then the S1 and S2 formed the conformation to recognize AFM1, decreasing the peak current (curve f).
The fabrication of the aptasensor was also evaluated by the impedance.The semicircle diameter was related to the electron transfer.The increase in semicircle diameter reflected the increase in the interfacial electrontransfer resistance.As shown in Figure S1B, compared with the bare GCE (curve a), the diameter decreased after the Au-rGO (curve b) coating, indicating a fast charge transfer (Zhang et al., 2019).Curve c showed that the diameter increased due to the immobilization of S1.After adding MCH, the uncovered sites were also blocked, leading to a further increase in diameter (curve d).Subsequently, a double-stranded formation of S1-CS1 increased the semicircle diameter (curve e).After adding AFM1 and S2, the diameter increased even further due to steric hindrances of charge transfer (curve f) (Hui et al., 2022).The results of EIS charac-   terization further validated the successful fabrication of the ratiometric electrochemical aptasensor.
Optimization of Detection Conditions.To obtain an electrochemical aptasensor with excellent performance, the ratio of methylene blue (MB) to ferrocene (Fc) peak current signal (I MB /I Fc ) was used as the evaluation index to optimize the concentration of S1, ionic strength, pH and incubation time with AFM1, as shown in Figure S2.With the increase of S1 concentration, I MB /I Fc increased and then decreased (Figure S2A) because a high concentration of S1 could cause steric hindrance between neighboring S1, obstructing the AFM1 binding, leading to a decrease in sensitivity (Ma et al., 2023).Therefore, 1 μM S1 was used for further experiments.Similarly, with the increase of Na + concentration, I MB /I Fc increased and then decreased (Figure S2B), indicating that a specific ionic strength was required for the split aptamer to recognize the target (Yang et al., 2021).When the ion concentration was 200 nM, the performance of the aptasensor was the best.However, the aptamer could not bend and recognize the AFM1 under an environment with high ionic strength.Also, pH is an important parameter to ensure aptasensor performance (Chen et al., 2022;Ding et al., 2020;Ho et al., 2019).The results of Figure S2C indicated that the split aptamer could better recognize the AFM1 due to conformation formed in a neutral environment.It also possessed the greatest rate of electron migration, resulting in optimal sensing outcomes (Guo et al., 2021).Previous research has also demonstrated that aptasensors exhibited an optimal performance under neutral conditions (Dou et al., 2023;Wang et al., 2019), showing a good agreement with our result.The results of the incubation time with AFM1 were also shown in Figure S2D.The I MB /I Fc increased for the first 60 min and then leveled off, illustrating that AFM1 binding reached saturation (Tan et al., 2021).Therefore, 60 min incubation time was used for subsequent experiments.
Sensitivity of the ratiometric electrochemical aptasensor.To evaluate the performance of the ratiometric electrochemical aptasensor, the SWV peak current in response to different concentrations of AFM1 was carried out under optimization conditions, as shown in Figure 5A.The peak current of Fc decreased, and that of MB increased correspondingly with the increasing concentration of AFM1.Such behavior was because the double-strand was destroyed, and CS1 was released from the electrode.MB probe was close to the electrode due to the formation of the S1-AFM1-S2 complex.As illustrated in Figure 5B, I MB /I Fc value showed a linear relationship with the concentration of AFM1 with a range of 0.03 μg L −1 -2.00 μg L −1 .The limit of detection was 0.015 μg L −1 .The ratiometric value showed a more comprehensive detection range and lower detection limit than the calibration plots for AFM1 using individual Fc or MB signals (Figure 5C).Moreover, compared with previously published works, this sensor showed a high sensitivity (Table S1).The precision for 3 replicate measurements of 0.8 μg L −1 AFM1 was 5.6% (relative standard deviation).
Selectivity and stability of the ratiometric electrochemical aptasensor.To demonstrate the selectivity of the ratiometric electrochemical aptasensor, AFM2, AFB1, AFG1, and AFG2 was chosen as the competitive structural analogs.It can be seen from Figure 6A and Figure S3A that at the same concentration (0.5 μg L −1 ), an apparent ratio of the AFM1 caused current change (I MB /I Fc ).Also, the signal ratio was similar to that caused by the mixture.However, other competitive analogs could cause minor current ratio changes.The results indicated that the ratiometric aptasensor based on split aptamer showed reasonable specificity for AFM1 and that structural analogs could not interfere with AFM1 detection.
To investigate the stability of the aptasensor, the performance of the ratiometric electrochemical aptasensor was evaluated after 1, 2, 3, and 4 weeks (Figure 6B and S3B).As shown in Figure 6B, the 23% decreased ratio signal was obtained after 4 weeks compared with the newly constructed aptasensor.This indicated that the aptasensor exhibited good stability.
Practicability of the ratiometric aptasensor.The milk was chosen as a representative sample to illustrate the practicability of the ratiometric aptasensor.Given that the protein may potentially interfere with the detection signals, we added 18 mM CaCl 2 for its elimination (Parker and Tothill, 2009).After centrifugation, the supernatant was used for the detection of AFM1.It can be seen from Figure S4 that the ratio signal exhibited a linear range from 0.2 μg L −1 to 1.00 μg L −1 , and the detection limit was 0.05 μg L −1 .Comparison with High Performance Liquid Chromatography (HPLC), the accuracy and reliability of the proposed analytical technique were further verified.As shown in Table S2, the performance of the proposed aptasensor was equivalent to that of the HPLC method, with statistically insignificant differences (P < 0.05) from the HPLC-derived results.The results validated the application of the ratiometric aptasensor for detecting AFM1 in milk samples.

CONCLUSIONS
In this work, a novel ratiometric electrochemical aptasensor was developed for the sensitive and selective detection of AFM1.The aptamer was split to improve recognition and labeled with redox reporter and Li et al.: Ratiometric electrochemical aptasensor… sulfydryl.In addition, the Au-rGO nanomaterial was utilized to improve electroconductibility and increase the electrode surface area.The aptasensor exhibited high sensitivity, selectivity, and stability.Moreover, the aptasensor also demonstrated a linear range in milk samples after simple pretreatment.This work provides a new approach to enhance the sensitivity of aptasensors and holds great promise for detecting other targets.
developed an electrochemical DNA biosensor based on tetrahedral DNA nanostructure and MXene (Ti3C2) nanosheets for the detection of gliotoxin.The amplified electrochemical signals could be produced resulting from the efficient and rapid binding of the target due to the unique configuration of tetrahedral DNA nanostructures.The split aptamer was more straightforward and cost-effective than constructing tetrahedral structures.Liu et al. (2020) reported a split aptamer sensing platform utilizing the single molecule photobleaching technique for highly sensitive and selective detection of theophylline.A dual-mode lateral flow biosensor incorporating visible-surface-enhanced Raman spectroscopy based on the split aptamer regulated CRISPR/Cas12a and gap-enhanced Raman tags was developed for 17beta-estradiol.Chen et al. (2022) developed a label-free aptasensor based on the fusion of a binary split G-quadruplex and the aptamer for the ultra-sensitive detection of malachite green.
Scheme 1. Schematic illustration of the ratiometric electrochemical aptasensor fabrication for detection of AFM1.

Figure 5 .
Figure 5. the SWV peak current in response to different concentrations of AFM1 (0-2.0 μg L −1 ) (A), a linear relationship on the concentration of AFM1 with a range of 0.03 μg L −1 -2.00 μg L −1 (B) and comparison with the calibration plots for AFM1 using individual Fc or MB signal (C).

Table 1 .
Li et al.: Ratiometric electrochemical aptasensor… All the oligonucleotides used in this work Note: bases marked in red are deliberate mismatched base pairs.