A novel fluorescent aptasensor based on mesoporous silica nanoparticles for the selective detection of sulfadiazine in edible tissue

2.1 Material and reagents

All standard antibiotics were purchased from Sigma-Aldrich (St. Louis, MO, USA). GO (2 mg/mL, 500 nm), streptavidin magnetic beads and carboxyl functionalized magnetic beads (MBs) (30 mg/mL, 2 × 100 beads/mL) were purchased from Invitrogen (Carlsbad, CA, USA). MSN-NH₂, 1-(3-Dimethylaminopropyl)-3-ethyl carbodiimide (EDC) was supplied by Sangon Biotech (Shanghai, China). Black polystyrene microtiter plates were purchased from Thermo Fisher Scientific Inc (Wilmington, USA). All other chemicals were of analytical grade and were obtained from Beijing Chemical Reagent Company (Beijing, China).

The primer sequences listed below have all been used in previous studies (Shi et al., 2020). All oligonucleotide sequences were synthesized by Shanghai Sangong Biotechnology Co., Ltd. (Shanghai, China). The ssDNA library for aptamer selection:

• 5′-FAM-GACAGGCAGGACACCGTAAC-N40-CTGCTACCTCCCTCCTCTTC-3′,

• Forward primer (FP):5′-FAM-GACAGGCAGGACACCGTAAC-3′,

• Reverse primer (RP):5′-biotin-GAAGAGGAGGGAGGTAGCAG-3′,

• cDNA: 5′-COOH-ATCCCATTGGGTT-3′.

2.2 In vitro selection of aptamers

The GO-SELEX technique was used to select aptamers in this study (Kou et al., 2020). Briefly, 200 μL of 500 nM ssDNA library was denatured at 95 °C for 10 min, quickly cooled on ice for 10 min and then incubated at 25 °C for 25 min. For the initial screening, 200 μL of 500 nM ssDNA library was incubated with an equal mole number of SDZ for 1 h at 25 °C under agitation. Subsequently, 2 mg/mL GO solution (mass ratio of ssDNA/GO was 1:25, Fig. S1) was added to the reaction solution and incubated at 25 °C under stirring for 20 min. ssDNA not bound to SDZ was adsorbed on the surface by GO via π-π stacking. After incubated for 20 min and centrifuged at 15,000 rpm for 10 min, the supernatant was collected.

The collected supernates were amplified with FP and RP. The original PCR reaction system was 2 × Taq Master Mix 100 μL, FP (50 μM) 4 μL, RP (50 μM) 4 μL and template solution 92 μL. A total of 11 cycles were perfomred, with the parameters set as follows: pre-denaturation at 95 °C for 5 min, at 95 °C for 30 s, at 58 °C for 30 s, and at 72 °C for 10 s, followed by the extension at 72 °C for 1 min (Fig. S2). The PCR amplification product was purified with streptavidin magnetic beads, and then the streptavidin magnetic beads were washed with NaOH (50 mM). The eluate was collected as the library for the next round of screening. Sulfonamides were then used as negative screening. Finally, SDZ was added for positive screening again and again until the ssDNA content in the system did not change (Fig. S3). UV–VIS spectroscopy (UV–VIS spectroscopy, Hitachi UH5300, Japan) was used to determine the ssDNA content at the end of each screening cycle.

2.3 Affinity characterization of sequence and sequence truncation analysis

First, SDZ dissolved in dimethylformyl was added to the EDC-activated MBs solution and incubated for 10 h with shaking.Subsequently, the MBs were rinsed three times with phosphate-buffered solution (PBS, PH = 7.4). Next, different concentrations of aptamers (25, 50, 100 150, 200 nM) labeled with FAM (λ_ex = 492 nm and λ_em = 518 nm) were mixed with MBs and incubated in the dark for 1 h. MB was washed with PBS for 3 times, then washed with 50 mM NaOH, and finally the supernatant after washing was collected. The fluorescence intensity in the supernatant was used to characterize the affinity of aptamers. The fluorescence intensity was measured with Vorioskan LUX (Thermo Scientific, USA). Affinity was expressed by calculating the dissociation constant (K_d) value. The K_d value was calculated according to the equation Y = B_max × X/(K_d + X), Origin 9.0 software was used for nonlinear regression analysis, where X is the concentration of aptamer, Y is the relative fluorescence intensity, and B_max is the maximum number of binding sites (Sadeghi et al., 2018, Kou et al., 2020, Zhu et al., 2021).

The candidate aptamer with the lowest K_d value was selected as the subsequent aptamer truncated analysis. Multiple sequence alignments were performed on the primary structure pairs of sequences using DNAMAN (https://www.lynnon.com/dnaman.html) to find conserved motifs. The secondary structures of these candidate aptamers were predicted at room temperature using internet-based Mfold (https://www.unafold.org/).

2.4 Molecular Docking

The tertiary structures of aptamer were obtained from RNAComposer. (https://rnacomposer.cs.put.poznan.pl/). The SDZ secondary structure was mapped using Chemoffice 2019 software and its energy was minimized. To determine the possible binding site between SDZ and SDZ30-1 (a truncated aptamer), molecular docking was performed using Autodock 1.5.6 (https://autodock.scripps.edu) and then visualized using PYMOL software (https://pymol.org).

2.5 MD trajectory analysis

MD simulations were performed using the GROMACS 2016.6 and G_MMPBSA software (Kumari et al., 2014, Autiero et al., 2018). Equilibration was performed at constant particle number, volume and temperature and isothermal isobaric. The aptamer-target complexes were first subjected to 100 ns of implicit solvent model MD simulations, followed by explicit solvent MD simulations of the resulting structurally stable complexes. The stability of aptamer-target complex was then assessed using the root mean square deviation (RMSD).

2.6 Combined with free energy calculations

The MM/PBSA method, which is also known as the Molecular Mechanics/Poisson-Boltzmann surface area for calculations, was used to estimate binding free energy and energy decomposition of MD trajectories. The binding free energy is split into a molecular mechanics term and a solvation energy which are calculated separately. The basic principle of the calculation is the difference between the binding free energy of two solvated molecules in the bound and free states or comparing the binding free energy of different solvated conformations of the same molecule (Klimovich et al., 2015). $$\mathrm{\Delta }{\mathrm{G}}_{\mathrm{b}\mathrm{i}\mathrm{n}\mathrm{d}}=\mathrm{\Delta }{\mathrm{G}}_{\mathrm{c}\mathrm{o}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}\mathrm{x}}-(\mathrm{\Delta }{\mathrm{G}}_{\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{e}\mathrm{p}\mathrm{t}\mathrm{o}\mathrm{r}}+\mathrm{\Delta }{\mathrm{G}}_{\mathrm{l}\mathrm{i}\mathrm{g}\mathrm{a}\mathrm{n}\mathrm{d}})$$ $$\mathrm{\Delta }{\text{G}}_{\text{bind}}=({\text{E}}_{\text{vdw}}+{\text{E}}_{\text{ele}})+({\text{G}}_{\text{polar}}+{\text{G}}_{\text{nonpolar}})$$ ΔG_bind is the free energy of binding, E_vdw is van der Waals energy, E_ele is the electrostatic energy in the gas phase, G_polar is the polar solvation energy, G_nonpolar is non-polar solvation energy.

2.7 Construction of the aptasensor

The schematic diagram of the fluorescent aptasensor constructed on silica was shown in Fig. 1. First, 20 mg of MSN-NH₂ was dispersed in PBS (PH = 7.4), 50 µL of carboxylated cDNA (10 µM) was activated with EDC, and then they were mixed and incubated for 2 h to form MSN-NH₂/cDNA complexes. 50 µL of aptamer (10 µM) was added and incubated in the dark for 3 h, and the pellet was retained by centrifugation to form the MSN-NH₂/cDNA/aptamer complex. Finally, the compounds were resuspended in PBS and stored at 4 °C for later use.

Close

Fig. 1

Schematic diagram of a fluorescent aptasensor based on MSN-NH₂ to detect SDZ.

Export to PPT

A standard curve was established as follows. 10 µL of the above particle suspension (aptamer concentration of 600 nM) was added to 150 µL of SDZ solutions of various concentrations (0, 3.125, 6.25, 12.5, 25, 50, 100 ng/mL) diluted in PBS. Then, the mixed solution was incubated for 1 h at 25° C in the dark. Finally, the supernatant was retained by centrifugation, and its fluorescence value was measured. The limit of detection (LOD) was calculated according to LOD = 3 SD/slope, SD was the standard deviation of the fluorescence intensity of the blank sample, and the slope was the slope of the linear regression response curve (Yan et al., 2022).

2.8 Aptasensor for analysis of SDZ in real-world samples

The aptasensor was validated using real-world samples spiked with known concentrations of SDZ (50, 100, and 150 µg/kg). The samples were purchased from the local market, and the processing method was referred to the supplementary information. The original samples were confirmed to be free of SDZ by HPLC. The spiked samples were tested using aptasensor and HPLC, and both recovery rate and coefficient of variation were evaluated. The recovery formula was as follows: (measured concentration/known concentration) × 100%. The concentration of each spiked recovered sample was measured five times.

3.1 Characterization of aptamers

All candidate sequences were divided into five families based on homology (Table S1). Based on the multiple sequence alignment of thirty candidate sequences, most of the candidate sequences shared the conserved motifs ‘TGG’ and ‘GGT’. Subsequently, the secondary structures of the candidate sequences were analyzed and the conserved motifs were located on the stem loop as important parameters for screening candidate aptamers. (Macdonald et al., 2016, Ye et al., 2019). Finally, SDZ1, SDZ15, SDZ22, SDZ25, and SDZ30 were selected as candidate aptamers for subsequent experiments (Fig. S5).

The K_d vaule of each candidate aptamer was then measured by fluorimetry to evaluate their affinity to SDZ (Ma et al., 2015). Targeted at a fixed amount of SDZ (50 ng/mL), the binding reaction was carried out with different concentrations of FAM-labeled candidate aptamers. The fluorescence intensity in the supernatant was evaluated after elution. The amount of aptamer used was proportional to the intensity of the fluorescence. A non-linear fit using Origin 9.0 software allowed the dissociation constants of each aptamer to be determined according to the equation Y = B_max × X/(K_d + X). As shown in Fig. 2A, the K_d values of SDZ15 (K_d = 245.42 nM) and SDZ30 (K_d = 217.77 nM) were significantly lower than other candidate aptamers. The specificity of SDZ30 and SDZ15 was then evaluated. The results were shown in Fig. 2B and Fig. S4, the relative fluorescence intensity obtained by SDZ30 combined with SDZ was much higher than that of other antibiotics, and SDZ15 also showed a certain preference for other sulfonamide antibiotics. This showed that the specificity of SDZ30 to SDZ was better than that of SDZ15. Therefore, SDZ30 was selected for subsequent experiments.

Close

Fig. 2

(A) The affinity of each aptamer was determined using a non-linear regression equation, and the relationship between FAM-aptamer concentration and relative fluorescence intensity was studied; (B) Verification of the specificity of SDZ30 to SDZ, (1) SDZ, (2) SN, (3) SMR, (4) SMZ, (5) Kanamycin, (6) Chloramphenicol, (7) Oxytetracycline, (8) Doxycycline, the aptamer concentration was 1 μM, and the concentration of all antibiotics was 100 ng/mL.

Export to PPT

3.2 Optimization of the aptamer sequence

To make the aptamer have a better practical application ability, the sequence of SDZ30 was truncated. The secondary structure of SDZ30 was shown in Fig. 3A, and the conserved motifs “TGG” and “GGT” were located in the two stem-loops, respectively. According to the result of previous studies, the binding sites of aptamers to targets were usually in stem-loops, G-quadruplexes, and pseudoknots (Chen et al., 2014, Macdonald et al., 2016, Wu et al., 2019, Gao et al., 2020). Therefore, we truncated the two stem loops and named them SDZ30-1 (Fig. 3B) and SDZ30-2 (Fig. 3C). As shown in Table 1, the affinity of SDZ30-1 (K_d = 65.72 nM) was approximately 4 times higher than that of SDZ30. The affinity of SDZ30-2 (K_d = 210.41 nM) remained unchanged from that of SDZ30. Therefore, we choose SDZ30-1 (5' – AACCCAATGGGAT − 3') as the aptamer for practical application.

Close

Fig. 3

SDZ30 secondary structures and truncated aptamer secondary structures.(A) SDZ30; (B) SDZ30-1; (C)SDZ30-2.

Export to PPT

3.3 Aptamer/ligand equilibrum simulation

MD simulations were carried out with SDZ30-1 using SDZ, sulfanilamide (SN), sulfamerazine (SMR) and sulfamethazine (SMZ) as ligands, respectively. Structural changes of the SDZ30-1/ligand complex during the simulation were observed using RMSD (Chen et al., 2019). As shown in Fig. 4, in the range of 0–20 ns, the RMSD value curves of the four systems fluctuate significantly, indicating that the nucleic acid backbone was moving violently during this process (Xie et al., 2020). The average RMSD was 0.5 nm within 20 to 100 ns, as shown in Fig. 4A, and there was no noticeable fluctuation, showing that the SDZ30-1/SDZ has stabilized. The SDZ30-1/SN was stable from 40 ns to 100 ns, with an average RMSD of 0.6 nm, showing that it was less stable than the SDZ30-1/SDZ (Fig. 4B.). Fig. 4C and D showed that the RMSD value curves of SDZ30-1/SMZ and SDZ30-1/SMR showed no signs of stability from 0 to 100 ns, indicating that the stability of SDZ30-1/SMZ and SDZ30-1/SMR was weaker than SDZ30-1/SN. This indicated that one of the reasons why SDZ30-1 specifically binds to SDZ was that SDZ30-1/SDZ has the highest stability.

Close

Fig. 4

RMSD values for SDZ30-1 with (A) SDZ, (B) SN, (C) SMR, (D) SMZ.

Export to PPT

3.4 Binding free energy analysis of SDZ30-1/ligand

To investigate the effect of aptamer/ligand binding free energy variations on SDZ30-1 affinity, the binding free energy of SDZ30-1 to SDZ, SN, SMR, and SMZ was calculated using the MM/PBSA method (Sponer et al., 2014). As shown in Table 2, the binding free energies of SDZ30-1 and the four sulfonamides were −17.37 Kcal/mol, −7.33 Kcal/mol, −4.87 Kcal/mol, and −3.43 Kcal/mol, respectively. As show in Fig. 4, the SDZ30-1/SDZ system was the most stable, with the highest electrostatic energy and the lowest solvation energy, while the SDZ-1/SMZ system was the most unstable, with the lowest electrostatic energy and the highest solvation energy. Therefore, the van der Waals energy and the electrostatic energy played a pivotal role in the specific binding of SDZ30-1, while the solvation energy can hinder the specific binding of SDZ30-1.

3.5 Structural analysis of SDZ30-1/ligand

As a representative structure for the SDZ30-1/ligand, we used the stable structure from the kinetic simulations. As shown in Fig. 5, the four ligand were located near the pocket formed by the conserved sequence “TGG” (The eighth, ninth, and tenth bases of the 5′ end of SDZ30-1: T8, G9, G10), and the distance between the four ligand molecules and the pocket was positively correlated with the size of molecule (V_SDZ = 233.7 A³, V_SN = 159.6 A³, V_SMR = 254.0 A³, V_SMZ = 269.1 A³) (Li et al., 2020b). In the structure diagram, the green bonds represent hydrogen bonds, the yellow bonds represent π-anions, the red bonds represent CH/π interactions, the blue bonds represent π-sulfur bonds, and the gray bonds represent π–π stacking. Among them, π–π stacking and CH/π interaction belong to van der Waals interaction, and hydrogen bond belongs to electrostatic interaction.

Close

Fig. 5

Three-dimensional structure of (A) SDZ30-1/SDZ, (B) SDZ30-1/SN, (C) SDZ30-1/SMR, (D) SDZ30-1/SMZ.

Export to PPT

For SDZ30-1/SN, as shown in Fig. 5B, SN entered the binding pocket due to its small size, and SN formed hydrogen bonds with G9 base and G10 base, respectively. However, because SN lacks a six-membered carbon–nitrogen ring, the binding free energy contribution of G9 base and G10 base to SDZ30-1/SN was lower than that of SDZ30-1/SDZ (Fig. S6). For the SDZ30-1/SDZ, as shown in Fig. 5A, there was a π-anion interaction between the T8 base and the six-membered carbon–nitrogen ring, forming a delocalized π bond with -NHSO₂-, so the binding pocket can accommodate SDZ well. There was a π-π stacking between G9 base and the benzene ring, with the shortest interaction distance of 4.12 Å. Thus, T8 base and G9 base contributed the most to the binding freedom of SDZ30-1/SDZ (Fig. S6). For SDZ30-1/SMR, as shown in Fig. 5C, due to the large size of the SMR molecule, it cannot fully enter the binding pocket, and only π-π stacking occured with the T8 base of the conserved motif, and the interaction distance was 4.9 Å. The additional methyl group of the SMR molecule increases the steric hindrance of SDZ30-1 binding to the SMR. For SDZ30-1/SMZ, Fig. 5D demonstrated that the two additional methyl groups in SMZ increase the steric hindrance to accessing the binding pocket, resulting in SMZ not interacting with the conserved motif “TGG”.

3.6 The principle of the aptasensor

The detection mechanism of the aptasensor was shown in Fig. 1. The carboxyl-modified cDNA was attached to MSN by forming an amide bond with MSN-NH₂. FAM-labeled aptamers were immobilized on the surface of MSN by Watson-Crick base pairing. When the presence of SDZ was detected in the sample, FAM-labeled aptamers left the surface of MSN and entered the supernatant to bind SDZ. The supernatant was then retained by centrifugation, and the fluorescence intensity in the supernatant was positively correlated with the concentration of SDZ. Finally, quantitative detection of SDZ can be achieved by the measurement of the fluorescence intensity.

3.7 Optimisation of the aptasensor conditions

The aptamer’s concentration and the aptamer/target’s incubation time were chosen as parameters to optimize the aptasensor. As shown in Fig. 6A and B, the concentration of the SDZ30-1 was 600 nM, and the incubation time of aptamer/target was 75 min for the best sensor performance. Under optimized experimental conditions, Fig. 6C showed a good linear relationship between fluorescence intensity and SDZ concentration in the dynamic range of 3.125 to 100 ng/mL (y = 0.02357x + 0.044, R² = 0.9841). And the LOD of aptasensor for SDZ was 1.68 ng/mL. Furthermore, this detection method was compared with previously reported methods. As shown in Table 3, the LOD of the proposed method was comparable to or even lower than some previously reported methods. Therefore, the fluorescent aptasensor detection method had the advantages of high sensitivity, easy operation, less time consuming and low cost, which can be used as a new method to detect SDZ residues.

Close

Fig. 6

(A) The influence of the incubation time of SDZ and aptamer on the fluorescence intensity; The concentration of SDZ was 50 ng/mL. (B) The influence of SDZ30-1 concentration on the fluorescence intensity; (C) linear fitting of the fluorescence intensity of the aptasensor; (D) The relative fluorescence intensity of other antibiotics was determined using the fluorescence intensity of SDZ as a standard (1) SDZ, (2) SN, (3) SMR, (4) SMZ, (5) Kanamycin, (6) Chloramphenicol, (7) Oxytetracycline, (8) Doxycycline.

Export to PPT

Under optimized experimental conditions, 50 ng/mL of several antibiotics (SDZ, SN, SMR, SMZ, Kanamycin, Chloramphenicol, Oxytetracycline, Doxycycline) were used to evaluate the specificity of the aptasensor for the detection of SDZ. As shown in Fig. 6D, the relative fluorescence intensity of other antibiotics and structural analogues of SDZ was less than 20%. These results indicated that the aptasensor has good specificity for detecting SDZ.

3.8 Accuracy of aptasensor in detecting real samples

To evaluate the accuracy of the aptasensor in detecting real-world samples, recovery experiments were performed on milk and honey. As shown in Table 4, the recoveries were 95.12–105.47%, and the coefficient of variation ranged from 3.34 to 13.18 %. It was confirmed that the accuracy and precision of the aptasensor were within an acceptable range. Furthermore, the results of aptasensor were positively associated (R² > 0.8687) with the results of HPLC detection results, indicating that the aptasensor was reliable in detecting SDZ.