Preparation and Validation of Ultra-sensitive Aptamers by Computer Simulation and Site-Specific Mutation

Aptamer is widely used in the detection field as a new type of probe due to their advantages of easy synthesis and modification. Due to the traditional aptamer screening method, the aptamers that are screened by SELEX technology generally have redundant sequences, which affect the binding affinity and detection sensitivity. In this study, a structural transformation strategy for rationally shortening was established by analyzing the secondary structure, finding the sequence of the binding site by base mutation, and double cloning the binding sites. We obtained a 27-mer new aptamer SEQ.A52 with the dissociation constants (Kd) of 10.74 nM, which is 21 times higher than the original aptamer. When used to detect enrofloxacin (ENR) with the fluorescent method, the sensitivity could reach 50 to approximately 1000 pM, and the limit of detections (LOD) was only 41.35 pM, which was 88 times higher than the original aptamer. All the results implied that the strategy could be used for structural modification of some other aptamers to increase affinity and sensitivity.


Introduction
Aptamer, which refers to single-stranded DNA or RNA artificially synthesized through in vitro systematic evolution of ligands through exponential enrichment (SELEX), is a new type of recognition probe. As an artificially synthesized molecule, aptamer has some unique advantages over antibodies in terms of molecular size, synthetic accessibility, and chemical modification, which make it can be widely applied in food safety and other research fields (Seok Kim et al. 2016).
High affinity and specificity are two key indicators to evaluate the performance of the unique sequence of aptamer against target (Kimoto et al. 2016). However, traditional aptamer screening technology SELEX is carried out with selection-inherent amplification procedures, which is prone to the formation of artificial by-products of PCR. The length of the screened aptamer is generally long and difficult to determine the binding effect of specific base sequence . Some previous studies have proved that the sequence length could significantly affect the sensitivity of the detection method based on aptamer (Han et al. 2014). This is also the reason why aptamer-based assays could not replace the standard antibody-based methods. Therefore, it is important to improve the sensitivity by investigating the combination of aptamer to target and screening some shorter aptamer containing the specific base sequence.
The spatial structure of an aptamer commonly consists of two parts, the nucleotides at its two terminals hybridize with each other to form a stem for its secondary structure stabilization , while the other nucleotides form an irregular loop to recognize the target, specifically and the substructures of the loop determined its binding affinity and detection sensitivity (Kwon et al. 2014).
Recently, more and more researchers tried to optimize the aptamer structure to improve the sensitivity and had reported that the detection methods established with truncated aptamers showed a lower limit of detection (LOD) and excellent sensitivity (Lv et al. 2018). This strategy generally increased the affinity between the aptamer with the target 10 times, but it could not achieve a greater order of magnitude improvement (Shi et al. 2013). At the same time, some people had proposed the use of macromolecule immobilization strategies to improve affinity (Huang et al. 2005). The incorporation of protein or other macro-molecular substance at some positions of aptamer can promote the formation of a G-quadruplex in the loop regions, which could apparently enhance the binding affinity with targets (Esposito et al. 2018), but this method was lack of universality and there was no regular method for the determination of macro-molecular substances. Therefore, it is of great importance to establish a new method to better optimize the structure of the aptamer and to improve the sensitivity and specificity of aptamer (Lee et al. 2018).
In our previous study, we obtained a 37-mer aptamer specific to ENR, and in this study, through computer simulation, the secondary structure of the aptamer and the energy level of bases which is binding to the target was obtained and the shortening assay was determined (Sha et al. 2021). Furthermore, through the strategy of base replacement, the structure of the aptamer was modified and the binding site was determined. And then the binding site was copied by cloning. After the determination of the dissociation constant (Kd) of the optimized aptamer to ENR, a new strategy for the optimization of the structure of the aptamer was established (Gooch et al. 2017).

Shortening and Screening of Aptamers
The secondary structures of the 37-mer aptamer (named SEQ.3) specific to ENR were simulated with software MFold (http:// www. unafo ld. org) based on the energy minimization principle to select the secondary structure that is most easily formed. All the aptamers were synthesized and their dissociation constants (Kd) to ENR were determined by equilibrium filtration assay. Briefly, the aptamers were dissolved and diluted with binding buffer (25 mM Tris-HCl buffer, pH 8.0 with 100 mM NaCl, 25 mM KCl, and 10 mM MgCl 2 ) to the concentration of 0-50 μM. One hundred microliters of ENR at the concentration of 50 μM was added and incubated at 37 °C for 1 h. Then the solution was centrifuged for 5 min at 12,000 × g in a 3000 Da ultrafiltration centrifugal tube the concentration of ENR in the filtrate was measured by UV-Vis at 278 nm (Akki et al. 2015). The Kd was calculated by nonlinear regression analysis using Eq. (1) and Origin software (Wang et al. 2007): where Y represents the degree of saturation, Bmax represents the maximum number of binding sites, and X represents the concentration of unbound ssDNA. The aptamer with the higher affinity was selected for structure mutation.

Aptamer Structure Mutation
Site-specific mutagenesis was carried out for the structure mutation of the selected aptamer according to the principle proposed by Kimoto M. et al. (2016). Two bases were a group and mutated in accordance with the principle of A-T base pair and G-C base pair replacement. A total of 7 new aptamers (SEQ.A1 to SEQ.A7) were obtained in a clockwise direction on the ring of the aptamer. After synthesis, the Kd constants of all the aptamers to ENR were determined as the method described in the "Shortening and Screening of Aptamers" section, and the aptamer with the highest affinity was chosen for binding site analysis and further modification. At the same time, in order to determine the accuracy of our sitedirected mutagenesis, we performed individual mutations on all single bases on the loop and determined the Kd value.

Identification, Modification, and Characterization of Aptamer
Combining the results of base mutations and using Autodock simulation docking results as verification, the docking site was finally determined and modified by double, triple, and quadruple clones on the binding sites with the TT bases as the linker arm (Slavkovic et al. 2020). The Kd constants of the modified aptamers were measured as the method described in the "Shortening and Screening of Aptamers" section, and the structural changes during binding were determined by the circular dichroism (CD) spectrum. Briefly, 500 nM aptamer and 500 nM ENR were mixed in 500 μL of binding buffer (25 mM Tris-HCl buffer, pH 8.0 with 100 mM NaCl, 25 mM KCl, and 10 mM MgCl 2 ) and incubated overnight at 37 °C (Mehlhorn et al. 2018). Then the CD spectra were measured by a Chirascan CD spectrometer instrument over the wavelength range from 200 to 400 nm scanned at 200 nm per min. The same concentration of aptamer without ENR incubated at the same condition was used as control.

Sensitivity and Specificity of the Detection with Aptamer to ENR
A fluorescent detection method for ENR based on aptamer after optimization was established as follows: 50 µL of 2 µM FAM-modified aptamer was mixed with 50 µL of ENR ranging from 0 to 1000 pM and incubated in MOPS buffer (containing 10 mM MgCl 2 , pH 7) for 1 h at 35 °C (Biniuri et al. 2018), then 100 µL of AuNPs was added. After another incubation of 20 min at 35 °C, the mixture was centrifuged with 14,000 × g for 10 min. The fluorescence intensities (F) of the supernatants were measured by fluorescence spectrophotometer with Em = 480 nm, Ex = 518 nm. The fluorescence intensity without ENR (F 0 ) was used as blank control and the correctional value ΔF (ΔF = F -F 0 ) was calculated to estimate the concentration of ENR.

Analysis and Shortening the Aptamer
The length of aptamers screened by traditional SELEX technology was generally longer about 60 bases. Some of these bases did not play a role in the binding to target but affected the affinity of the aptamer (Jia et al. 2020). In this study, when simulated with software MFold based on the energy minimization principle, we have obtained a secondary structure that requires the least energy to form, that is, the easiest to form under target induction. The secondary structure of the SEQ.3 aptamer consisted of four parts: two central loops (A and B), a stem connecting the loops, and a non-connecting stem ( Fig. 1A) (Yan et al. 2019). It was obtained by software, base energy when binding to the target, which from low to high was labeled as purple, blue, green, yellow, orange, and red, and showed that loop A was the main binding domain according to the principle that the bases with higher binding energy were more   Alsager et al., in which the affinity of 35-mer aptamer cut from an original length of the 75-mer aptamer based on structure simulation was increased by 2.5 times (Alsager 2015). At the same time, this result showed that with the help of secondary structure simulation and base binding energy analysis, the key domains of aptamer recognition could be effectively predicted, which provided a basis for further mutation transformation to improve the recognition performance.

Aptamer Structure Mutation
Every base in the aptamer sequence played a certain role in the binding of the aptamer to the target. When the base pair changed, the affinity of the aptamer also changed due to the difference in chemical forces such as hydrogen bonds and van der Waals forces (Kaiser et al. 2018). We followed the principle of DNA mutation on the loop and mutated in accordance with the principle of A-T base pair and G-C base pair replacement. The sequences of 7 new aptamers and their Kd values were listed in Table 1. Among these aptamers, except SEQ.A1 was mutated from the stem, the other 6 sequences were mutated in the loop ( Fig. 2A).
Something interesting was that when replaced CG with TA, including CC with TT (SEQ.A4) and GG with AA (SEQ.A2, SEQ.A3, and SEQ.A7) in the loop, the affinity of the new aptamer was reduced. But when replaced AT Structure simulation showed that the secondary structure of SEQ.A5 had not changed, but the base binding energy had changed obviously. The original low-energy A-T had been transformed into a high-energy G-C (Fig. 2B). This indicated that the overall base binding energy of aptamer affected the affinity of the aptamer with the target (Liu et al. 2020). When the aptamer was combined with the target, a certain three-dimensional structure was formed by twisting and folding. In this process, the bases with high binding energy on the ring formed a bond-fixing structure, which overcame the flexibility of the aptamer and thus obtained better binding performance (Yang et al. 2017).
At the same time, in order to ensure that other replacement methods did not produce better affinity and further determined the binding site, we replaced a single base on the loop according to the replacement method of A-T base pair and G-C base pair replacement, thereby further verifying our results. The Kd value was shown in Table 2, and the result was consistent with the double base substitution. When the base positions of SEQ.A5 and SEQ.A6 were substituted, the Kd value was the lowest. And, it provided a basis for us to determine the binding site later.

Identification, Modification, and Characterization of Aptamer
Autodock docking simulation has been used to determine the binding site of aptamer successfully (Zhou et al. 2019). After site-directed mutation, the affinity of the aptamer was increased significantly, but the binding sites were not clear. When simulated with Autodock 4.2 software, it showed that the ENR was mainly combined with G11, C12, and G15 through the action of hydrogen bonds (Fig. 3), so the five bases from G11 to G15 were identified as the binding sites. Among these bases, G11 and C12 were just the mutation sites in the site-directed mutation, which further proved the correctness of the site-directed mutation.
The five key bases (GCCAG) were cloned twice, three times, and four times with TT base as the connecting arm. Their new aptamers (SEQ.A52, SEQ.A52, and SEQ.A54) were obtained and their Kd values were determined. As shown in Table 3, the double-cloned aptamer had the lowest Kd value, which was only 11.07 nM. After triple cloning, the aptamer was 35-mer with the Kd value of 224.05 nM, which was similar to that of the original SEQ.3. When cloned four times, the length of the poly-clonal was too long and the aptamers interfered with the formation of steric binding sites, resulting in a decrease in affinity. Structure simulation showed in Fig. 4 that the secondary structure of SEQ. A52 was different from the original SEQ.3. Although it still had two central circles, the stem length at the end of the SEQ.A52 circle was asymmetrical and the base difference between the two circles was not as large as that of SEQ.3 (Zhu et al. 2017). It was speculated that after double cloning, the binding site changed from one to two, thereby binding more targets and improving the affinity. CD spectroscopy was used to determine the secondary structure changes of the aptamer during the combination with ENR (Okazawa et al. 2000). As shown in Fig. 5A, the original SEQ.3 presented the standard B-type aptamer absorption peaks formed of single-stranded DNA, with the maximum and minimum absorption characteristic peaks at 280 nm and 245 nm, respectively, which was consistent with the results reported by Nagatoishi et al. (2007). After shortening, the characteristics of SEQ.A showed an obvious blue shift. The positive peak and negative peak absorption peaks were 270 and 240 nm, respectively. This might be due to the change in the length of the aptamer, resulting in the lack of the central circle of the B region in the secondary structure (Fig. 1A). As the site-directed mutant of SEQ.A, SEQ.A5 presented similar characteristics to SEQ.A, but SEQ.A52 showed a significantly different spectrum. The negative peak could not be found to shift to 235 nm and the positive peak at 270 nm was split into two peaks, a high peak at 260 nm and a low one at 290 nm, which belonged to the process intermediate from type B aptamer to type A aptamer . At the same time, the result was similar to the secondary structure simulation result, except that the secondary structure of SEQ.A52 was quite different from the secondary structure of SEQ.3, with two large and one small circle in the middle. It was precisely because of the difference in the secondary structure that it had better binding properties.
In addition to the CD comparison of the four aptamers, ENR was also added to observe the changes in the secondary structure of the aptamers. after combination, the peak values of SEQ.3, SEQ.A, and SEQ.A5 were decreased due to the interaction between the base stack and the spiral superstructure, the higher the affinity of aptamer, the more obvious the decrease of peak value, so SEQ.A5 had the most obvious effect because it had the highest affinity (Fig. 5D). To SEQ.A52, except for the peak reduction, the positive peak at 260 nm also showed a split state, indicating that it is binding to the target was different from other types (Ye and Yin, 2008). This was also the reason for its high affinity with the target.

Sensitivity and Specificity of the Aptamer
The fluorescent detection method for ENR with aptamer was established based on the principle that the aptamers were adsorbed on the surface of AuNPs through electrostatic interaction and induced the quenching effect of AuNPs on fluorescence. When ENR was not present, the aptamer modified with FAM was adsorbed on the surface of AuNPs CD of SEQ.A5 with or without ENR (E). CD of SEQ.A52 with or without ENR ◂ and caused fluorescence quenching. When ENR was present, the aptamer competed from the AuNPs surface, and the fluorescence value was restored and could be quantitatively analyzed with the concentration of ENR (Song et al. 2019).
The detection results with SEQ.3, SEQ.A, SEQ.A5, and SEQ.A52 were shown in Fig. 6. All the fluorescence values had a good linear correlation with the increase of ENR concentration. SEQ.A52 presented the highest sensitivity, in which the linear correlation was fitted as y = 0.2302x -2.2942. The correlation coefficient was 0.9928 in the low-concentration range of 50 to 1000 pM. The LOD was calculated as 41.35 pM by the concentration corresponding to the fluorescence value at three times the standard deviation of 10 blank samples without ENR (Fig. 6A). This result was 100 times more sensitive than the fluorescence method established using SEQ.3, and the LOD was reduced by about 88 times (Fig. 6B). At the same time, the sensitivity has been greatly improved compared with the shortened aptamer (SEQ.A) and mutated aptamer (SEQ.A5) shown in Fig. 6C and D.
Specificity results (Fig. 6E) showed that there were no significant changes between SEQ.A52 and SEQ.3 to the all the analogues except CIP, and the cross-reaction rate was less than 15%. But to CIP, SEQ.A52 presented lower crossreactivity than SEQ.3, and the reason might be that the double cloning of the binding site to ENR in SEQ.A52 decreased the recognition to CIP. All the results indicated that with the optimization of the aptamer structure, reasonable shortening, rational modification through site-directed mutagenesis, and the use of docking simulation to find the binding site while reasonably doubling could greatly improve the binding property of the aptamer (Tan et al. 2013).

Conclusion
In the study, a method for the screening of ultra-high sensitivity aptamer was established. Through the structure simulation and shortening of existing aptamer SEQ.3 to ENR, the site-directed mutation was used to find the binding site. A new aptamer SEQ.A 52 was obtained with double cloned the binding site. Compared to the original aptamer, SEQ.A52 had a shorter sequence and much stronger affinity. When used to detect ENR with the fluorescent detection method, the LOD was only 41.35 pM, which was 88 times higher than the original aptamer. All the results implied that the strategy could be used for structural modification of some other aptamers to increase affinity and sensitivity.