SELEX Procedure. In this study, GO-SELEX was carried out as an enrichment process for selecting aptamers against ENR and OFLO. GO-SELEX is a method that allows the aptamers and small molecules to interact in solution without any hindrance. The experiment design was based on the principle that the free ssDNA can adsorb onto the GO surface by π-π stacking and hydrogen bonding interactions. This circumstance provides the opportunity to remove ssDNA’s from the solution to eliminating those or collecting for incubating them with a target. As both ssDNA and GO have a negative charge, the adsorption can take place in a buffer with an appropriate salt concentration. The prior studies showed the adsorption of ssDNA on GO surface in such a buffer can just be accomplished in a minute48,49.
The used binding buffer contained 100 mM NaCl, 5 mM MgCl2, and 20 mM Tris-HCl at pH 7.4. The solution with the mentioned ionic strength and pH mimics the biological condition that not only leads to the adoption of a stable spatial structure by aptamer but also this condition facilitates the adsorption of ssDNA on GO Sheets. It should be noted that the zwitterionic antibiotics ENR with pka1 5.88 and pka2 7.74 and OFLO with pka1 6.06 and pka2 8.22 exist in a neutral zwitterionic form in water and they do not bear any charge in that pH.50 The applied GO-SELEX was operated in three styles to finding the selective and specific aptamers against ENR and OFLO. The used three styles are demonstrated in Fig. 1.
In the initial cycles of the SELEX procedure, by first incubating the aptamer and antibiotic and then adding GO to the solution, some oligo strands that had a chance to have interaction with antibiotic molecules were separated from the huge variety of strands in the random pool. The saturation percentage in the first three cycles of SELEX was between 4–9% (Figure SI-2). As in the initial cycles, all of the selection parameters were constant the saturation values did not fluctuate out of the above range. The low variation in the saturation values could arise from the efficiency of the ethanol precipitation method.
In the middle of the screening process, the counter selection was done, in which some of the antibiotics with similar molecular structures to the main antibiotics were incubated with the sub-pool, then the strands that did not have an affinity to the counter molecules were collected by GO. This process aimed to discard the strands that could be bound to similar structures. Different antibiotics were used as the counter molecules. One of them was ciprofloxacin that is a member of quinoline family antibiotics and it is structurally identical to ENR except for an ethyl group on the piperazinyl region. Other counter antibiotics were florfenicol and Sulfabenzamide that belong to organo-fluorine and sulfonamide families, respectively.
In the final stages of selection, the denatured pool was first incubated with GO. Having washing GO with binding buffer, the target molecules were incubated with the washed GO together with the adsorbed ssDNA’s on its surface. Under this condition, only strands that have a high affinity to antibiotic molecules detach from the GO surface, and by refolding they interact with antibiotic molecules. It can be said that in some cases the affinity to the target predominates over the interaction with GO, by nearing the target to the oligo strands on the GO surface, the interaction between these two molecules induced refolding of aptamer and results in detaching of the strand from the GO surface48,49. This process of selection has some advantages. By performing this process, the dsDNA that may exist in the solution would be eliminated and the only strands that have a high affinity to the target will be selected.
From the results (Figure SI-2) it can be seen that in SELEX cycles the most fluctuation in saturation values is contributed to changing the selection strategy or increasing the pressure on the enrichment process like descending the aptamer concentration or shortening the incubation time.
Cloning and Sequencing. The selected sequences at the end of the SELEX procedure were cloned by Thermo Scientific PCR Cloning Kit. Having extracted the plasmid, two different PCR reactions were carried out to pick the right strands for sequencing. These PCR reactions were performed in two series, one with universal M13 primers and another with SELEX primers. In the former, it was expected to see the PCR products with 227 base pairs and in the latter with the SELEX primers, it was expected to see the PCR products with 72 residues. The product of these two kinds of PCR was run on 1% agarose gel and those plasmids that had two different and specific bands in front of the 100 and 300 base pairs band of the ladder, were selected for sequencing. The procedure is demonstrated in Fig. 2. The image of the agarose gel is represented in Figure SI-1B.
The NA sequence of the selected aptamers associated with the quinolone antibiotics is presented in table SI-1. The aptamers related to Enrofloxacin and Ofloxacin were named ENR-APT and OFLO-APT, respectively.
The analysis of the random part related to each aptamer was shown that the random part of ENR-APT and OFLO-APT contained 57.5% and 72.5% GC residues, respectively. Both aptamers were analyzed using the Pairwise Sequence Alignment online software from European Bioinformatics Institute (EMBL-EBI) site. The result of this analysis demonstrated that there was a 38.1% similarity in the paired format between the random parts of two aptamers (Table SI-2).
Analyzing the Secondary Structure of Aptamers. The secondary structure of both aptamers was illustrated in figure SI-3. The predicted Gibb’s free energy for ENR-APT and OFLO-APT were − 16.68 kcal.mole-1 and − 20.30 kcal.mole-1, respectively. The predicted secondary structure of both ENR-APT and OFLO-Apt have the stem-loop (hairpin) patterns that two and three bulges can be seen in the stem part of ENR-APT and OFLO-Apt, respectively. In both aptamer’s secondary structures two mismatches between G and T can be seen in the stem region and both of them have a single strand segment at the 5’ and 3’ end.
Dissociation Constant (K d ). To evaluate the level of dissociation tendency of aptamer-antibiotic complex, the equilibrium filtration method was carried out. This assay was performed in the light of the assumption that the ratio of aptamer-antibiotic binding in solution is 1:1. The calculated Kd were 12.025 ± 0.007 µM and 6.98 ± 2.44 µM, corresponding to aptamer-ENR and aptamer-OFLO complexes, respectively. The saturation binding curve plot is represented in Fig. 3.
Analyzing Binding Assay of Aptamers for Quinolone Antibiotics. The result of the specificity binding experiments showed that both aptamers had a similar affinity to the three fluoroquinolone family antibiotics; enrofloxacin, ciprofloxacin, and ofloxacin. This result is depicted in Fig. 4. By contrast, both aptamers recognized the quinolone antibiotics from the sulfabenzamide and florfenicol.
As it is demonstrated in Fig. 5 all the three quinolone antibiotics have a quinolinemonocarboxilic scaffold and all the functional groups that interact with aptamers (e.g. the fluoroquinolone scaffold and piperazinyl region) in the used quinolone antibiotics are identical, then it is not far from the mind that both aptamers showed affinity to all of them. Also, the indiscrimination of both aptamers showed that the bulky ethyl and methyl groups do not affect the interaction between antibiotics and aptamers. This evidence clarified that most probably the complementarity in shape has not an effective role in the interaction between the isolated aptamer and its target.
Despite large biomolecules (like; proteins, etc.) that have high diversity in their functional groups which result in very specific interaction with aptamers, the lack of various epitopes in small molecules leads to lower affinity and lower specificity between them and the aptamers14. There are several reported cases in which the introduced aptamer could not distinguish between structurally very similar small molecules39,51−55. For instance, Stojanovic et al. have reported an aptamer that could not distinguish between some cinchona alkaloids. They raised the hypothesis that the aromatic motif is the main recognized part of these small molecules therefore their aptamer responded to alkaloid structures with a similar aromatic region53. In another study, Gülbakan et al. presented a cocaine aptamer that could bound to another alkaloid even stronger than its main target. Regarding another aptamer related to the adenosine, they represented that this aptamer had the affinity to several analogs of adenosine with the same hydroxyl group on the sugar ring. They expressed that the prior studies confirmed the important role of these hydroxyl groups in the interaction with aptamer39.
So far in this study, it has been proposed that indiscrimination of aptamers between the quinolone antibiotics arises from the similarity of antibiotic structures. But it should be noted that the alignment between the sequences of two aptamers revealed that there is a 38% similarity in their nucleotides. Another hypothesis that was brought up by this fact, is the similar nucleotides of both aptamers could be located in the binding site. But this needs further investigation to find out which part of the aptamer has the most important role in the interaction with antibiotics. Since both isolated aptamers can distinguish quinolone antibiotics from the other class of antibiotics significantly, they have the potential to be used as the recognition element in apta-sensors to monitor the amount of this family of antibiotics in the different matrices.
Investigation of the Aptamer- Antibiotic Binding Using UV-Vis Spectroscopy. To study more about the antibiotic-aptamer complex and to trace changing the small molecule's optical properties, an UV-Vis spectroscopy assay was performed. In this regards, a solution of each antibiotic in the absence of aptamer and in complex with aptamer was prepared and the ɛ value of both ENR and OFLO was determined. The comparison of the obtained ɛ value related ENR and OFLO, in the absence and in the interaction with aptamer, revealed that 41.98 ± 0.18 % and 22.45 ± 0.40 % reduction happened in ɛ value associated to complex formation between ENR and OFLO with their related aptamers, respectively. And also 3 and 1 nm red shift was observed in λmax of ENR and OFLO, respectively (Fig. 6). This finding confirmed that the aromatic region of both antibiotics was involved in the interaction with the aptamer strand.
The aromatic region, carboxylic group, carbonyl, and flour of quinolone antibiotics, are placed on a flat surface and the flat region of small molecules can be located in a coplanar position with the aptamer bases. As a result of this coplanar position, they can interact with each other by π-π stacking 56. As a consequence of this interaction, less energy is needed for electronic excitation so hypochromic effect and bathochromic shift appear in small molecule’s UV-Vis spectrum. Previous studies claimed that in some cases the extinction coefficient of the small molecule could be reduced 40%-60%36,57.
Investigation of the Aptamer- Antibiotic Complex Formation Using High-Resolution Mass Spectrometry. All the advantages of mass spectrometry that were mentioned before, required us to apply this method for more investigation into the formation of the aptamer-antibiotic complex. In this regard, each aptamer was incubated with its related antibiotic in ammonium acetate buffer with pH 7. Even though the monovalent ions (Na+) and divalent ions (Mg+ 2) in binding buffer help in adopting the spatial folding of oligonucleotides by decreasing the electrostatic repulsion between the phosphates groups in the polyanionic backbone of oligonucleotides strands, they also cause the aggregation of DNA strands in the droplets formed in the ion source which results in the suppression of oligonucleotides peaks in the spectrum. For this reason, ammonium acetate is used instead of others salts. In the gas phase, ammonium ions transfer one proton to oligo strands and become neutral species in form of the gas. By proton transferring, the salt interfering is eliminated moreover results in neutralizing some phosphate groups in oligonucleotide strands and also by reducing the negative charge and decreasing repulsion between phosphate groups, the oligo strands can adopt the stable 3D structure39,40,58.
The recorded MS spectra are shown in Figs. 7 and 8. In MS spectra related to Aptamer-ENR and Aptamer-OFLO complexes, 22 and 27 peaks with wide charge distribution were identified, respectively (tables SI-4 and SI-5). All the recognized peaks were at the ratio of 1:1 between aptamer and antibiotics.
Gülbakan et al. made the hypothesis that the aptamer strands with less negative charge have more impact spatial structure in the gas phase and this could result in the formation of a more stable complex with the ligand39. But our obtained results were not inconsistent with their hypothesis and several peaks corresponding to aptamer-antibiotic complexes with a wide range of charge distribution were observed in mass spectra. The charge distribution of complexes is represented in figure SI-4.
Keller et al. previously presented some evidence stating that in the vacuum and in the absence of solvent competition, the hydrogen bonding and electrostatic interaction between aptamer and target become stronger. So in the aptamer-target complex where the H-bonding and electrostatic interaction is the predominant kind of interaction, they are more stable in the gas phase. But the π-π interaction would be the last non-covalent bonding that survives in the gas phase. In general, the kind of interaction between small molecules and aptamers will determine the stability of their complex in the gas phase41. As in both obtained mass spectra of Aptamr-ENR and Aptamer-OFLO complexes, several peaks were identified. It can be concluded that the formed complexes were stable in the gas phase and this stability in the gas phase can reflect the contribution of H-bonding or electrostatic interaction between aptamer and antibiotic.
The obtained evidence by specificity test, UV-Vis spectroscopy, and mass spectrometry suggested the predominant interaction between the isolated aptamer and the quinolone antibiotics are H-bonding, π-π interaction, and electrostatic interaction. In our opinion, every study and research about the interaction between aptamers and small molecules can lead to wider usage of the science of aptamers and help design more efficient Apta-sensors to be used for enhancing the quality of human health and even in environmental protection.