Molecular Crowding Modulates SARS‐CoV‐2 Aptamer Affinity

SARS‐CoV‐2 aptamer is a favorable candidate for the recognition and detection of SARS‐CoV‐2, owing to its small size and easy synthesis. However, the issue of compromised binding affinities in real samples and targeting mutant SARS‐CoV‐2 hinder wide applications of the aptamer. In this study, it is discovered that molecular crowding could increase binding affinity of CoV2‐6C3 aptamer against RBD (Receptor Binding Domain) of SARS‐CoV‐2 via increasing the absolute value of the enthalpy change. The values of the equilibrium dissociation constant in molecular crowding decrease by 70% and 150%, respectively, against wild‐type and mutant RBD compared with those in buffer without crowding. Moreover, the detection limit of SARS‐CoV‐2 pseudovirus is up to 5 times lower under molecular crowding compared to dilute conditions. The discovery deepens the understanding of aptamer‐target interaction mechanisms in crowding conditions and provides an effective way to apply SARS‐CoV‐2 aptamer for virus recognition and detection.


Introduction
Aptamers are functional affinity handles that evolved in vivo and in vitro. [1][4][5] Because of these advantages, aptamers serve as attractive alternatives to antibodies for molecular recognition. [6]However, aptamers generally have substantially lower affinities than antibodies in complex systems.This intrinsic weakness of aptamers is due to the limited chemical diversity of natural nucleic acids (A, T/U, G, and C). [7]10] Thus, for aptamers to reach their full potential as recognition molecules, their binding affinities in complex biological environment must be evaluated and maximized (lower equilibrium dissociation constant, K d ).
[13] The chemically modified aptamers can be obtained by de novo evolution or substitution.However, most chemically modified nucleotides are not efficiently recognized by traditional DNA polymerases, limiting amplification and sequencing of the aptamer candidates.On the other hand, the strategy of substituting natural bases by chemically modified bases at specific locations requires considerable trial and error to identify suitable bases among the aptamer sequence. [14,15]Therefore, a convenient method to improve aptamer affinity still critically needs to be developed.
Previously, we discovered an aptamer (CoV2-6C3) against the RBD (Receptor Binding Domain) of SARS-CoV-2. [16]In followup experiments, we found that the binding affinity and maximum number of binding sites (B max ) of CoV2-6C3 aptamer are better in plasma than in buffer (Figure 1C and S1, Supporting Information).Specifically, the values of K d and B max of CoV2-6C3 aptamer in plasma are 36 nM and 469, while CoV2-6C3 aptamer in buffer are 51 nM and 258.[22] However, positive or negative effects and determinant factors on binding affinity seem to be different among different aptamers. [23]Herein, curiosity ultimately drove us to take a further look at the reasons of the CoV2-6C3 aptamer binding improvements in plasma Scheme 1. DOI: 10.1002/sstr.202300089SARS-CoV-2 aptamer is a favorable candidate for the recognition and detection of SARS-CoV-2, owing to its small size and easy synthesis.However, the issue of compromised binding affinities in real samples and targeting mutant SARS-CoV-2 hinder wide applications of the aptamer.In this study, it is discovered that molecular crowding could increase binding affinity of CoV2-6C3 aptamer against RBD (Receptor Binding Domain) of SARS-CoV-2 via increasing the absolute value of the enthalpy change.The values of the equilibrium dissociation constant in molecular crowding decrease by 70% and 150%, respectively, against wild-type and mutant RBD compared with those in buffer without crowding.Moreover, the detection limit of SARS-CoV-2 pseudovirus is up to 5 times lower under molecular crowding compared to dilute conditions.The discovery deepens the understanding of aptamer-target interaction mechanisms in crowding conditions and provides an effective way to apply SARS-CoV-2 aptamer for virus recognition and detection.

Effect of PEG on CoV2-6C3 Aptamer Binding
Polyethylene glycol (PEG), a neutral cosolute with high water solubility, low toxicity, biodegradability, and inertness to nucleotides, is widely used to mimic molecular crowding conditions. [24,25]The addition of high-molecular-weight PEG increases the viscosity of the solution and the effect of excluded volume, [26] which was conducive to the stability of nucleic acid structure. [27]As a result, PEG was applied as the macromolecule crowding agent in this work.We characterized the binding performance of CoV2-6C3 aptamer against SARS-CoV-2 RBD in different concentrations of PEG 4000 (Figure 1A).When the concentration of PEG increased from 0% to 16% w/v by volume, the ratio of CoV2-6C3 binding fluorescence signal to that of a random sequence gradually became higher.When the concentration of PEG exceeded 16% w/v, aptamer binding dropped steeply, probably because the overcrowding condition limited the collision of the aptamers and targets.In 16% w/v PEG, the binding ratio increase of CoV2-6C3 reached a plateau, about 1.3 times better than that without PEG.Therefore, we chose 16% w/v PEG as the subsequent simulated molecular crowded microenvironment.In addition, CoV2-6C3 aptamer in 16% w/v PEG did not recognize the RBD/spike protein of other coronaviruses such as SARS-CoV and MERS-CoV, while maintaining the same selectivity as CoV2-6C3 aptamer in binding buffer without PEG (Figure 1B).
K d values of CoV2-6C3 against RBD protein in 0% and 16% w/v PEG were evaluated.Compared to dilute conditions, for each crowded microenvironment, better binding affinity CoV2-6C3 aptamer was exhibited, with K d decreasing by %1.7-fold (from 51 to 30 nM).Moreover, the B max value of aptamer-RBD binding in 16% w/v PEG was 2.3 times greater than that in 0% w/v PEG, indicating that there are more interaction events of RBD and CoV2-6C3 in 16% w/v PEG than that in dilute conditions (Figure 1C).These results in 16% w/v PEG (30 AE 4 nM) were similar to the values in human plasma (36 AE 7 nM, Figure S1, Supporting Infromation), indicating that aptamer-RBD binding under crowding conditions is similar to in vitro conditions.
Next, the effect of molecular crowding on the thermodynamics of CoV2-6C3 binding was characterized by determining the standard changes in enthalpy (ΔH ⊖ ), entropy (ΔS ⊖ ), and free energy (ΔG ⊖ ) during the binding reaction (Figure 1D).To obtain the values of ΔH ⊖ , ΔS ⊖ , and ΔG ⊖ , the K d values of CoV2-6C3 at different temperatures were measured by flow cytometry and further analyzed by the van't Hoff equation (Figure S2 and S3, Supporting Information). [19]The calculated values of ΔH ⊖ and ΔS ⊖ indicate that the crowding condition causes the interactions of the CoV2-6C3 with RBD to change from entropy-driven (ΔH ⊖ = À4.4 kcal mol À1 , TΔS ⊖ = À5.4 kcal mol À1 ) to enthalpy-driven process (ΔH ⊖ = À7.7 kcal mol À1 , TΔS ⊖ = À2.5 kcal mol À1 ).This phenomenon may be caused by the conversion of molecular crowding into a confinement effect, which decreases the degree of CoV2-6C3 freedom.In addition, the crowding condition increased ΔG ⊖ by 0.4 kcal mol À1 (À9.8 and À10.2 kcal mol À1 in dilute and crowding environments, respectively), suggesting that molecular crowding increases the formation of specific binding, such as hydrogen bonds, electromagnetic effects, and van der Waals forces.

Effect of Molecular Crowding on CoV2-6C3 Aptamer Structure
To characterize the effect of molecular crowding on the structural change of CoV2-6C3 aptamer, several CoV2-6C3 probes with the same sequences and different fluorescent labeling sites were synthesized.Given that CoV2-6C3 aptamer consists of two loops, Loop 1 (in orange) and Loop 2 (in purple), the fluorophore (FAM) and quencher (BHQ1) were placed at each stem-loop edge of CoV2-6C3, respectively (Figure 2A-C).Compared to CoV2-6C3-Loop1 and CoV2-6C3-Loop2 in 0% w/v PEG, there was a 2.7 and 1.6-fold increase of fluorescence intensity in 16% w/v PEG, respectively (Figure 2A,B).Since PEG itself does not have fluorescence (Figure S4, Supporting Information), these results indicate that the fluorophore is farther from the quencher, and Loop1 or Loop2 are in a more "open" state in the condition of molecular crowding.When the RBD was introduced on the basis of CoV2-6C3 in 0% w/v PEG and 16% w/v PEG, the fluorescence intensity of CoV2-6C3-Loop1 and CoV2-6C3-Loop2 further increased (Figure 2A,B), with higher values than in the same conditions without RBD.These fluorescence increases indicated that the aptamer conformation in CoV2-6C3-RBD interaction is more likely to have the loops in a relatively open state.Furthermore, the variation of the overall conformation of CoV2-6C3 was also investigated, which showed the same trend as Loop1 and Loop2 in the presence of RBD or crowded condition (Figure 2C).These results show that the structural change of Loop1 and Loop2 tends to the favorable structural state, which interacts with the target under molecular crowded conditions.In addition, the different position of the fluorescent label had almost no effect on aptamer binding to RBD (Figure S5, Supporting Information).
To better understand the aptamer-RBD interactions under different environments, we carried out molecular docking and molecular dynamics simulations.First, molecular dockings were performed to predict the structure of CoV2-6C3 in buffer and molecular crowding conditions (Figure 2D).The predicted stem ends of the two loops are further apart in the crowded environment than in a buffer solution (Figure S6, Supporting Information), which are similar to the experimental results (Figure 2C).In addition, the predicted binding model of RBD and CoV2-6C3 aptamer in molecular crowding conditions contains larger consecutive binding interfaces than that in dilute conditions, resulting in a more negative Gibbs free energy change in crowding conditions than that in buffer solution (Figure 2D).

CoV2-6C3 Aptamer Binding for Mutant RBD in Molecular Crowding
As an RNA virus, SARS-CoV-2 tends to form multiple mutant strains, possibly leading to reduction or failure in the binding ability of aptamers evolved against wild-type SARS-CoV-2.Therefore, the binding affinities of CoV2-6C3 against mutant RBD protein with and without molecular crowding were also determined.As shown in Figure 3A, CoV2-6C3 aptamer could still recognize the RBD protein with K417N: E484K: N501Y mutations, and the signal-to-background ratio was still the highin 16% w/v PEG.K d value of CoV2-6C3 against mutant RBD protein (K417N: E484K: N501Y) was evaluated to be 18 AE 2 nM in 16% w/v PEG, 2.6 times lower than that in dilute conditions (Figure 3B).Moreover, the B max value of CoV2-6C3 against mutant RBD protein (K417N: E484K: N501Y) in 16% w/v PEG is 1.3 times larger than that in 0% PEG (Figure 3B).These results indicated that the binding affinity CoV2-6C3 aptamer under molecular crowding conditions is better than that in buffer.Thus, molecular crowding helps CoV2-6C3 aptamer to resist SARS-CoV-2 mutation escape.

Aptamer-Based Viral Detection in Molecular Crowding
We next investigated the potential of using the crowding condition strategy for aptamer-based SARS-CoV-2 detection.In this sandwich assay, the biotinylated CoV2-6C3 aptamer was immobilized on streptavidin-beads and then complexed with CoV2-6C3 aptamer that is linked to HRP (horseradish peroxidase) enzyme.Detection is accomplished by assessing the conjugated HRP activity via incubation with the substrate ABTS to produce a green color (Figure 4A).As a result, the green color of the reaction mixture gradually deepens with increasing SARS-CoV-2 pseudovirus concentration.As expected, due to increased aptamer affinity in a crowding environment, compared to detection in buffer, the color change was more pronounced in molecular crowding conditions for the same concentration of pseudovirus (Figure 4B).As shown in Figure 4C, there is a linear relationship between the absorbance value at 405 nm and the pseudovirus concentration in the range of 1-100 pM, and the estimated limit of detection (LOD) is 1.18 pM in 16% w/v PEG, which corresponds to 7.1 Â 10 8 virus particles per mL.Compared with the LOD in 0% w/v PEG, the detection sensitivity increased by %5 times.The result suggests that the detection sensitivity can be significantly improved by the addition of a suitable dose of cosolute to form a crowding reaction system.In the future, the sensitivity can be further improved by combining with other signal output methods.

Conclusion
In summary, we discovered that molecular crowding improved the binding affinity of CoV2-6C3 aptamer against SARS-CoV-2 RBD.The increase in binding affinity of the aptamer in a crowded environment is due mainly to the increase in the absolute value of the enthalpy change, essentially forming more favorable interaction between the aptamer and RBD.The enhancement of the aptamer affinity not only improves the sensitivity of aptamer-based SARS-CoV-2 pseudovirus detection, but also improves the broad-spectrum binding ability on SARS-CoV-2 mutants.We expect this work to shed light on the aptamer-target interaction mechanism in crowding conditions and promote the development of applying SARS-CoV-2 aptamer for virus detection.
Peripheral Blood Samples: Peripheral blood samples were approved by the Human Research Ethics Committee at the First Affiliated Hospital of Xiamen University (Project number: KYX-2018-006).Blood samples were centrifuged 3000 rpm for 15 min to remove the blood cells.The supernatant was the plasma for experiments.
Flow Cytometry Analysis: To assess the binding affinity of the CoV2-6C3 in dilute and crowding conditions, positive RBD-Ni-beads or RBD (K417N: E484K: N501Y)-Ni-beads were incubated with 200 nM FAM-labeled CoV2-6C3 in different concentrations of PEG at 25 °C for 30 min.The beads were washed and suspended in 100 μL binding buffer.The fluorescence intensity of beads was measured by flow cytometry (FACSVerse, BD).The result of binding affinity was calculated by the equation (the fluorescence signal of the sample with the aptamer incubation: background signal)/(the fluorescence signal of the sample without the random sequence incubation -background signal).Measurements were repeated as n = 3 independent replicates and the error bar was calculated by the following formula.
To obtain the K d value of the aptamers in dilute and crowding conditions, a series of different concentrations of the CoV2-6C3 with fluorescent labeling was incubated with RBD-Ni-beads or RBD (K417N: E484K: N501Y)-Ni-beads.After 30 min incubation and washing with binding buffer, the mean fluorescence intensities analyzed by flow cytometry were recorded as Y and fit to the Equation Y = B max * X/(K d þ X ) using Origin 8.0 software.The experiments for obtaining K d were repeated as n = 3 independent replicates.
In the thermodynamic characterization assay, the equilibrium dissociation of the CoV2-6C3 aptamer at different temperatures was determined and van't Hoff analysis was further conducted.To measure the K d values of aptamers, a series of different concentrations of the CoV2-6C3 with fluorescent labeling was incubated with RBD-Ni-beads at different temperatures in dilute and crowded conditions.On the basis of the obtained K d values, the following equations were used to calculate the thermodynamic parameters, including the standard Gibbs free energy (ΔG ⊖ ), enthalpy (ΔH ⊖ ), and entropy (ΔS ⊖ ): 1).ΔG ⊖ = -RTlnK ⊖ and 2).ΔG ⊖ = ΔH ⊖ ÀTΔS ⊖ , where K ⊖ is the equilibrium constant, K ⊖ and K d are reciprocal, R is the gas constant, and T is the temperature in Kelvin.On the basis of Equation (1) and values of K d , the values of ΔG ⊖ can be calculated.Therefore, we measured the enthalpic and entropic contribution in different reaction conditions.
Fluorescence Spectrum Analysis: Double-labeled aptamer was prepared at 0.2 μM in 0% and 16% w/v PEG, and the labeled aptamers were denatured by heating at 95 °C for 10 min and then cooling to room temperature.Next, spectral scanning was carried out on a fluorescence spectrophotometer (FluoroMax-4, Horiba).All samples were excited at 488 nm, and the fluorescence emission spectra were recorded from 500 to 580 nm with 5 nm bandpass in a 200 μL quartz cuvette.The fluorescence intensities of the donor (FAM) and acceptor (BHQ1) were measured using an appropriate filter set.Image processing was performed with Origin9.
Molecular Docking and Dynamic Simulations: We searched the S protein structure of SARS-CoV-2 from the RCSB PDB data bank (http://www.rcsb.org,ID: 6VSB).We used mfold web server (http://mfold.rna.albany.edu/?=mfold) to predict the secondary structure of aptamer.Using the predicted secondary structure of the aptamer as a starting point, the corresponding 3D structures of the equivalent ssRNAs were then modeled and visualized in RNAcomposer.Molecular docking was performed with Rosetta after obtaining the 3D structures of aptamers and their target.The Amber FF99SB and AMBER PARM99 force fields were used for the protein and aptamer system, respectively.The final average structure was obtained based on 5000 snapshots, which were extracted from the last 10 ns trajectory of MDS using the Gromacs 5.1 software.
Colorimetric Assay: This experiment was performed on streptavidin (SA) beads.We first sealed SA beads overnight with 5% skimmed milk to reduce non-specific adsorption.Then, they were washed three times with washing buffer and 500 nM biotinylated CoV2-6C3 aptamers were added in binding buffer (including dilute conditions and crowding conditions) for 30 min.After repeating the washing procedure, different concentrations of the pseudotype virus of SARS-CoV-2 in binding buffer were incubated for 1 h.Next, 500 nM biotinylated CoV2-6C3 in binding buffer was added again for 30 min.Finally, the beads were resuspended in 100 μL binding buffer with streptavidin-conjugated HRP (1:1000 dilution) for 30 min.In the presence of H 2 O 2 , ABTS substrate was incubated for 15 min.The catalytic reaction was terminated by 2 M H 2 SO 4 .After the color reaction was complete, the SA beads were removed, and a plate reader was used to measure the absorbance at 405 nm.Measurements were repeated as n = 3 independent replicates.In the range of virus concentration from 1 to 100 pM, the linear relationship was fit with virus concentration as the horizontal coordinate and OD value as the vertical coordinate (all minus the background signal value).LOD value was calculated according to the formula LOD = 3 * SD/k, where SD is the standard deviation of the blank, and k is the slope value of the linear equation.

Figure 1 .
Figure 1.Flow cytometry to investigate the binding ratio of A) the aptamer CoV2-6C3 against RBD and B) selectivity study of the aptamer CoV2-6C3 against SARS-CoV and MERS-CoV in the presence and absence of PEG.RS represents the fluorescence intensity of random sequence (n = 3, mean AE s.d.).C) Dissociation constant of aptamer CoV2-6C3 against RBD and the maximum binding site of RBD in 0% (yellow) and 16% w/v PEG (blue).D) At 295 K, the thermodynamic parameters for the interaction of CoV2-6C3 with RBD protein in dilute conditions (left) and in molecular crowding conditions (right), respectively.

Figure 2 .
Figure 2. A-C) Fluorescence spectra of aptamers CoV2-6C3 with different fluorescent label positions to analyze structural changes.D) The results of molecular docking and dynamics simulations.The overall structure of CoV2-6C3-RBD in dilute conditions (left) and CoV2-6C3-RBD in molecular crowding conditions (right).

Figure 4 .
Figure 4. A) Schematic of the aptamer-based colorimetric sandwich assay for detection of pseudovirus in 0% and 16% w/v PEG.B) Photograph of the aptamer-captured pseudovirus at different concentrations in 0% (above) and 16% w/v PEG (below) via the colorimetric assay.C) Linear relationship between absorbance at 405 nm of the reaction solutions and pseudovirus concentration (n = 3, mean AE s.d.).