Selection of Novel DNA Aptamers against HPV16 Biomarker, E6: An In Silico Approach

Abstract

HPV16 E6 oncoprotein is a critical driver of cervical cancer, necessitating high-specificity diagnostic and therapeutic agents. This study presents an in silico selection of DNA aptamers targeting the HPV16 E6 oncoprotein through a pipeline of structure-based virtual screening and 250 ns Molecular Dynamics (MD) simulations, performed with GROMACS 2019 and the CHARMM36M. Using AutoDock VinaXB and HADDOCK, eight candidates were evaluated against the F2 RNA benchmark. MM/PBSA calculations identified Apt 9 (−143.76 kcal/mol) as the most potent lead, surpassing the affinity of the F2 control (−142.80 kcal/mol). Apt 8 emerged as a key consensus lead, exhibiting superior structural stability (RMSD 0.23 nm) and the most robust interaction network (11.49 hydrogen bonds). Notably, these leads primarily target critical Histidine residues (His78, His118, and His126) on the E6 protein. These results designate Apt 9 and Apt 8 as high-priority candidates for experimental validation, providing a robust theoretical foundation for novel HPV16-targeted interventions.

1. Introduction

Human papillomavirus type 16 (HPV16) stands as a primary etiological agent in the development of various malignancies, most notably cervical cancer, but also including oropharyngeal, anal, and other anogenital cancers. , Despite the implementation of widespread vaccination programs, HPV-driven malignancies remain a significant global health challenge, further compounded by the limited availability of targeted therapies for existing HPV-related tumors. The virus infects the basal cells of the epithelium and establishes a quiescent infection in the proliferative cells. Upon entry, the viral DNA persists as episomes, present in approximately 50–100 copies per cell, with early viral gene expression occurring at low levels. The urgent need for novel therapeutic strategies and early detection of biomarkers necessitates a deeper understanding of the molecular mechanisms underlying HPV infection and carcinogenesis.

While both HPV E6 and E7 oncoproteins are recognized as indispensable for maintaining the malignant state in HPV-positive cancers, rendering them attractive therapeutic targets, the E6 protein often presents a more compelling and advantageous target, particularly for ligand-based strategies like aptamers. E6 primarily drives malignant progression by facilitating the degradation of the tumor suppressor protein p53 via the ubiquitin-proteasome system and activating telomerase, which collectively contribute to immortalization, increased DNA damage, and cancer development. The continuous involvement of E6 in p53 degradation and telomerase activation is pivotal for the sustained immortality and genetic instability characteristic of established cancers, thereby making its inhibition a direct intervention against a core mechanism of malignancy. In contrast, the primary role of E7 is in the early stages of carcinogenesis, where it inactivates retinoblastoma family proteins, resulting in uncontrolled cell proliferation. Furthermore, E6 exhibits superior druggability and specificity, as its interactions with cellular targets like E6AP and p53 involve well-defined binding interfaces, such as the characterized E6AP binding pocket, which offers clear structural motifs amenable to ligand recognition. , Among the various classes of molecular ligands capable of targeting these distinct interfaces, nucleic acid aptamers have emerged as a particularly promising modality. Current HPV 16 treatments primarily focus on preventing infection through vaccination or managing associated lesions and cancers, as there is currently no direct antiviral cure for the infection itself. − Advanced therapeutic strategies, many still under investigation, often target the viral E6 and E7 oncoproteins, which are critical for HPV’s oncogenicity, using approaches like therapeutic vaccines, gene silencing, and small molecule inhibitors. − The proposed aptamer-based method, which aims to block HPV entry by specifically targeting the E6 protein, aligns with these cutting-edge approaches by offering a novel, targeted intervention that could potentially bridge the gap in direct antiviral treatment.

Aptamers, single-stranded (ss) oligonucleotides that bind to specific target molecules with high affinity and specificity, have emerged as promising candidates for targeted therapy. Their unique ability to fold into complex three-dimensional (3D) structures allows them to recognize a wide range of targets, including proteins, peptides, carbohydrates, and even whole cells. Aptamers can be readily synthesized, modified, and conjugated to various therapeutic agents, making them ideal for targeted drug delivery and imaging applications. The development of aptamer-based therapeutics for HPV16 infection holds great promise, as they can selectively target HPV16-infected cells, deliver therapeutic payloads, and minimize off-target effects. The versatility of aptamers enables therapeutic intervention by targeting critical stages in the HPV life cycle. Aptamers are smaller and can easily penetrate the tumor core. They also exhibit high thermodynamic stability and lack immunogenicity, facilitating safe and effective retention in target cells in vivo. Moreover, aptamers can be synthesized in vitro independently of biological systems, thus eliminating the potential risk of bacterial or viral contamination, and importantly, they are flexible for structural and chemical modifications, eventually extending their clinical applications. Furthermore, owing to the advantages and unique adaptability of aptamers for point-of-care platforms, aptamer technology has established a stable niche in in vitro diagnostics by enhancing the speed and accuracy of diagnoses. Aptamers may also be used in aptamer-based versions of immunoassays, including enzyme-linked oligonucleotide assay, Western blot, immunohistochemistry, and flow cytometry. These aptamer-based assays (or sensors) could exhibit enhanced sensitivity, specificity, and stability, improving the detection of particular pathogens. −

Given the immense potential of aptamers for HPV 16 diagnosis and therapy, efficient and rational design strategies are paramount. In silico methodologies, such as molecular docking and molecular dynamics (MD) simulations, offer powerful and cost-effective approaches to predict aptamer-target interactions, screen potential candidates, and refine their binding properties before extensive experimental validation. , Molecular docking, a widely utilized computational technique, rapidly predicts the binding orientation and affinity of a ligand, such as an aptamer, to a target protein. − This computational pipeline represents a significant advancement in the development of targeted therapies for HPV16-driven cancers, offering a novel and highly efficient approach to aptamer discovery against the E6 oncoprotein, as it allows for the virtual screening of numerous potential aptamer sequences against a protein of interest, conserving resources and significantly accelerating the discovery process. , While docking provides initial insights into binding modes, molecular dynamics simulations offer a more comprehensive understanding of the dynamic stability and conformational changes of aptamer-target complexes over time. , MD simulations are crucial for refining aptamer designs by assessing the strength and stability of interactions, thus enabling the selection of optimal candidates. Therefore, this study focused on the in silico selection of candidate aptamers for the diagnosis and therapy of HPV 16. These candidate aptamers were analyzed with MD simulation parameters to identify and select the top-performing candidate aptamers. To validate our computational pipeline and provide a benchmark for our novel DNA aptamers, we included the F2 RNA aptamer, a previously characterized E6-binding sequence from the literature, as a positive control. This allows for a direct comparison between our newly designed candidates and an established binder in terms of structural stability and binding energetics

2. Methodology

2.1. Materials and Software

The following software and Web server platforms were used in the present study: (i) AutoDock VinaXB (version 1.2.3, http://vina.scripps.edu/) (ii) MGLTools 1.5.4, (iii) PyMOL Molecular Graphics System (version 3.1.4.1), (iv) GROMACS 2019 package (v) CHARMM-GUI server. All simulations were performed using the computational resources provided by the Centre for High Performance Computing (CHPC), South Africa (https://www.chpc.ac.za/).

2.2. Structure Search, Validation, and Preparation

The HPV-16 oncoprotein-specific candidate aptamers were generated using an in silico approach. The 3D structure of the protein HPV-16 E6 (PDB ID: 4XR8) was downloaded from the Protein Data Bank database (RCSB PDB). 4XR8 is key as it captures the E6-E6AP-p53 ternary complex, directly targeting the oncogenic mechanism of E6. Targeting the E6 interface is a potent therapeutic strategy due to its crucial role in p53 degradation and viral specificity. Aptamers are well-suited to disrupt this interaction by binding to critical interfaces and blocking complex formation. The monomeric structure of the 4XR8 was validated using the SAVES web server (https://saves.mbi.ucla.edu/) tools, including ERRAT, WHATCHECK, and PROCHECK. The tools enable a comprehensive evaluation of structural integrity, stereochemical consistency, and sequence-structure compatibility. The initial validation utilized the ERRAT tool, which assesses crystallographic precision by identifying regions prone to potential errors based on electron density analysis, thereby providing a global measure of model quality. Subsequently, the WHATCHECK tool was applied to inspect stereochemical properties and chemical integrity, flagging any structural anomalies that could indicate deviations from expected biochemical properties. Additionally, PROCHECK analysis, which generates Chi1-Chi2 plots to examine side-chain torsion angles, offers a detailed assessment of local structure quality and highlights any outliers that might suggest deviations from expected structural norms. PROCHECK also generated Ramachandran plots to evaluate backbone dihedral angles, allowing for further stereochemical verification and confirming the overall conformational plausibility of the models. Together, these validation tools established a robust foundation for subsequent in silico analyses of 4XR8 as a potential diagnostic target. The structural quality of the 3D model of the HPV-16 E6 oncoprotein (PDB ID: 4XR8) was also assessed using the Swiss model-QMEAN web server (https://swissmodel.expasy.org/qmean), a composite scoring function that combines multiple structural descriptors to describe the major geometrical aspects of protein structures. QMEAN is designed to identify the best model within an ensemble of alternatives by assessing parameters such as torsion angle potential for local geometry, solvation potential, secondary structure agreement, and solvent accessibility agreement.

2.3. Binding Site Prediction Using DogSite Scorer and Virtual Screening via AutoDock Vina

The structure of the 4XR8 was subjected to the Dogsite scorer tool within the protein plus web server (accessible at https://proteins.plus/pages/about) to predict the active sites. The protein plus web server enables users to predict potential binding sites, identify similar binding sites for ensemble docking, and perform molecular docking of small molecules of interest into a binding site. This process requires access to and knowledge of a high number of tools with a multitude of parameters.

The active sites identified were used to create a grid box using MGL AutoDock Tools version 1.5.7 (accessible at https://ccsb.scripps.edu/mgltools/downloads/). Docking search spaces were defined by setting grid boxes to capture active and surrounding regions of the proteins. The grid boxes were set around the active novel amino acids. The grid box dimensions were set to x: 126, y: 100, and z: 126, centered at coordinates x: 1.64, y: 0.555, and z: −8.383. These parameters were selected to ensure adequate coverage of the protein surfaces, enabling comprehensive exploration of potential binding sites.

Briefly, the water molecules were deleted to streamline computational efficiency, as they are generally not involved in direct ligand-protein binding interactions. Kollman charges (−58.591) were added to approximate the electrostatic environment of the proteins. These charges enhance the accuracy of electrostatic potential calculations, a critical factor in docking predictions. Hydrogen atoms were retained at default settings to simulate physiological conditions and ensure the structure remains biologically relevant. Nonpolar hydrogens were merged in both structures to reduce computational load, given their minimal contribution to electrostatic interactions since they are attached to the nonelectronegative carbon atoms. Histidine (His) hydrogens were edited (His30, His67, and His68) to improve protonation accuracy. These modifications followed established protocols to enhance binding reliability through appropriate histidine protonation. Any missing atoms in the protein structures were repaired to maintain structural fidelity throughout the docking process. −

The prepared 4XR8 model, along with the 3D models of the proprietary candidate aptamers were submitted to the Centre for High Performance Computing (CHPC) stationed in Rosebank, Cape Town (South Africa), for virtual screening using the AutodockVinaXB tool (version 1.2.5). Autodock VinaXB is a docking tool used to predict the binding of receptor macromolecules to small molecules, such as ligands, and to determine bound conformations and binding energies through binding scores. AutoDock VinaXB is optimized for flexible docking, allowing both ligand and receptor flexibility to enhance the reliability of binding interaction predictions. This computational approach provided a robust platform for identifying high-affinity candidate aptamers against the 4XR8 model.

2.4. Design of Full-Length Candidate Aptamers, Molecular Docking Using High Ambiguity Driven Protein–Protein DOCKing (HADDOCK) Web Server and Interaction Analysis

The top-ranked candidate aptamers identified through virtual screening were subjected to full-length candidate aptamer design. Full-length candidate aptamers were designed by incorporating rigid structures on both ends. A total of eight candidate sequences (Apt1–Apt9) were initially processed; however, Apt4 was excluded from further analysis as it failed to yield a converged 3D structural model during the RNA Composer prediction phase. Consequently, only the eight successfully modeled candidates proceeded to the docking stage.In addition to the novel DNA sequences, the F2 RNA aptamer (5′-ATTCAACATTCGAGGTGGATGCTACGAATCAAC-3′) was incorporated as a positive control. The 3D structure of F2 was modeled using RNA Composer and subjected to the same 250 ns Molecular Dynamics production run and MM/PBSA analysis as the DNA candidates to ensure consistent benchmarking of the scoring methodology.

Molecular docking of the designed candidate aptamers against the 4XR8 protein was performed using the High Ambiguity Driven protein–protein DOCKing (HADDOCK) web server. HADDOCK is an information-driven flexible docking approach for the modeling of biomolecular complexes. HADDOCK requires the definition of ambiguous interaction restraints to guide the docking process. The ambiguous interaction restraints were defined based on the amino acid residues located at the active site of the protein and the interacting residues in the candidate aptamer, as identified from the Autodock Vina results. The HADDOCK web server employs a sophisticated algorithm integrating physics-based energy functions with empirical data to refine docking poses, improving accuracy in predicting binding orientations. HADDOCK scores consider various energy terms, including van der Waals interactions, electrostatics, desolvation, and restraint violation energy, providing a comprehensive assessment of the binding affinity and stability of the complex. For each complex, the top 10 models based on HADDOCK score were selected for further analysis. The top-ranked docked complexes were analyzed for binding interactions using Discovery Studio Visualizer software. This software facilitates detailed examination of hydrogen bonds, hydrophobic contacts, and other noncovalent interactions, providing insights into the molecular mechanisms driving aptamer-protein recognition and binding affinity.

2.5. MD Simulation Using GROningen Machine for Chemical Simulations (GROMACS)

MD simulations were performed for the top-ranked candidate aptamer-4XR8 complex to investigate its stability and dynamics over time. GROMACS 2019 was used to run 250 ns MD simulations of the candidate aptamer-4XR8 complex. The GROMACS package is a versatile tool for simulating the dynamics of biomolecular systems, providing detailed insights into conformational changes and atomic-level interactions. The complex was solvated in a cubic box with explicit TIP3P water molecules, ensuring proper hydration of the system. The system was neutralized by adding counterions (Na+ or Cl−) to maintain electrical neutrality. The system was minimized using the steepest descent algorithm to remove steric clashes and optimize the initial structure, ensuring a stable starting point for subsequent simulations. Following energy minimization, the system was equilibrated under NVT (constant number of particles, volume, and temperature) and NPT (constant number of particles, pressure, and temperature) ensembles for 250 ns each to equilibrate the temperature and pressure.

During the NVT equilibration, the temperature of the system was maintained at 310 K using the V-rescale thermostat, ensuring stable thermal conditions for subsequent simulations. In the NPT equilibration, the pressure was maintained at 1 bar using the Parrinello–Rahman barostat, allowing the system to adjust its volume to achieve equilibrium under constant pressure conditions. After equilibration, a 250 ns production MD simulation was performed with a time step of 2 fs under NPT conditions. The Particle Mesh Ewald algorithm was used for long-range electrostatic interactions, ensuring accurate treatment of electrostatic forces in the simulation. Periodic boundary conditions were applied in all directions to mimic an infinite system and avoid edge effects. The root-mean-square deviation (RMSD), radius of gyration (R _g), Solvent accessible surface area (SASA), root-mean-square fluctuation (RMSF), and hydrogen bonds (HB) in the range of 0–250 (ns) were calculated from the MD trajectories using GROMACS tools. The root-mean-square deviation measures the average deviation of the protein backbone atoms from a reference structure, providing insights into the overall stability and conformational changes of the protein during the simulation. RMSF quantifies the average fluctuation of each residue in the protein, highlighting regions of high flexibility and potential conformational changes. The hydrogen bond analysis provides information on the number and stability of hydrogen bonds formed between the candidate aptamer and protein, which are crucial for maintaining the integrity and binding affinity of the protein. The simulation setup and parameters were carefully chosen to ensure the accuracy and reliability of the MD simulations, providing valuable insights into the dynamic behavior of the aptamer-4XR8 complex. MD simulations offer a detailed approach to assessing the flexibility of both ligands and proteins by tracking the movement of individual atoms within the molecular field. MD simulations emerge as refined tools, addressing the limitations of traditional methods by modeling the time evolution of molecules through solving classical equations of motion, thereby capturing important details of the dynamic process. The accuracy of MD simulations depends on the force field used to approximate interatomic interactions, which is generally fitted to quantum-mechanical calculations and experimental measurements.

2.6. MM/PBSA Binding Free Energy Analysis

The binding free energy of the protein–aptamer complex was quantified using the Molecular Mechanics/Poisson–Boltzmann Surface Area (MM/PBSA) approach. This comprehensive strategy assesses the binding free energies of protein–ligand complexes by decomposing the total binding free energy into various energy components. Calculations were performed using the gmx_MMPBSA tool (version 1.4.3), which is integrated with GROMACS and implements the AMBER MMPBSA.py script. For the analysis, snapshots were extracted from the molecular dynamics production trajectory. , The calculation utilized frames from startframe = 1 to endframe = 500, with an interval = 10, meaning every 10th frame within this range was sampled. This resulted in 50 snapshots being uniformly extracted from the trajectory. This segment was chosen after confirming the structural stability and convergence of key trajectory metrics, such as the Root Mean Square Deviation of the complex and the binding free energy values themselves.

The binding free energy, denoted as ΔG, was calculated using the fundamental thermodynamic relationship

$$\mathrm{\Delta }G=\mathrm{\Delta }H-T\mathrm{\Delta }S$$

Here, ΔG signifies the change in binding free energy, ΔH represents the change in enthalpy, and TΔS is the entropic contribution at absolute temperature. This thermodynamic relationship provides crucial insights into the molecular interactions driving complex formation. The change in entropy (ΔS) in the MM/PBSA framework was evaluated through Normal Mode Analysis. This process involved minimizing the selected structures from the MD simulations to remove high-energy conformations that do not represent the typical energy landscape of the system. Subsequently, the vibrational frequencies were computed for the complex, the receptor (protein), and the ligand (aptamer) in their bound and unbound states. From these vibrational frequencies, the entropy was then calculated using statistical mechanics principles with the formula

$$\mathrm{\Delta }S=S_{\mathrm{complex}}-S_{\mathrm{receptor}}+S_{\mathrm{ligand}}$$

2.7. Principal Component Analysis

Principal Component Analysis was employed as a statistical method to simplify complex data sets by separating biologically significant protein domain movements from less significant localized atomic motions. As a linear transformation method, PCA identifies key movements within the data using a covariance matrix derived from atomic coordinates, which represent the degrees of freedom of the protein and aptamer residues. This approach transforms the data into a new coordinate system, where the most significant variance is aligned with the first principal component, and the second most significant variance is aligned with the second coordinate. A covariance matrix was generated using gmx_mpi covar (part of the GROMACS molecular dynamics package) to examine how atomic positions deviate from their mean in relation to one another. The eigenvectors and eigenvalues of the covariance matrix were then computed to identify the principal components. The eigenvectors (of the covariance matrix) were used to indicate the directions of the most significant variance, while the eigenvalues (which were obtained from the eigenvalue.xvg files and are the coefficients associated with eigenvectors) were used to indicate the amount of variance each principal component accounted for.

Finally, a 2D graph was created to visualize the distribution of variance along the two main principal component axes. This provided insights into the differences in protein movements throughout the simulation landscape. The percentage contribution of each eigenvalue to the total variance was calculated using the following formulas

$$\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{c}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{a}\mathrm{g}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{v}\mathrm{a}\mathrm{r}\mathrm{i}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{e}\mathrm{f}\mathrm{o}\mathrm{r}\mathrm{t}\mathrm{h}\mathrm{e}1\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{e}\mathrm{n}\mathrm{v}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{e}=\left(\frac{\mathrm{E}\mathrm{i}\mathrm{g}\mathrm{e}\mathrm{n}\mathrm{v}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{e}1}{\sum _{i=1}^N\mathrm{E}\mathrm{i}\mathrm{g}\mathrm{e}\mathrm{n}\mathrm{v}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{e}i}\right)\times 100\%$$

$$\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{c}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{a}\mathrm{g}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{v}\mathrm{a}\mathrm{r}\mathrm{i}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{e}\mathrm{f}\mathrm{o}\mathrm{r}\mathrm{t}\mathrm{h}\mathrm{e}2\mathrm{n}\mathrm{d}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{e}\mathrm{n}\mathrm{v}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{e}=\left(\frac{\mathrm{E}\mathrm{i}\mathrm{g}\mathrm{e}\mathrm{n}\mathrm{v}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{e}2}{\sum _{i=1}^N\mathrm{E}\mathrm{i}\mathrm{g}\mathrm{e}\mathrm{n}\mathrm{v}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{e}i}\right)\times 100\%$$

Where eigenvalue i for the i-th principal component, and N is the total number of eigenvalues, a combined percentage of variance from the first two principal components greater than or equal to 50% was considered indicative of these components capturing the predominant motions of the system.

3. Results and Discussion

E6 is one of the viral oncoproteins that contribute to the development of HPV-associated cancer. It targets the tumor cell of the host suppressor protein p53, which binds to it and promotes its degradation through the Ubiquitin-proteasome system. E6 also interacts with other cellular proteins and affects cell signaling pathways. To identify the target protein on HPV-16, the literature was reviewed.

3.1. 4XR8 PDB Structure Protein–Protein Interaction Analysis

The red-highlighted binding sites represent the direct points of contact between Chain D and Chain F. The amino acid residues in Figure are hypothesized to be critical determinants of this interaction. These residues are likely involved in forming various types of molecular bonds, such as hydrogen bonds, salt bridges, hydrophobic interactions, and van der Waals forces, which collectively contribute to the affinity and specificity of the binding. Arginine, Glutamic acid, Aspartic acid, Lysine, and Histidine can form salt bridges (electrostatic interactions between oppositely charged side chains) and hydrogen bonds. These interactions are strong and highly directional, contributing significantly to binding specificity and affinity. They are often found enriched at protein-binding sites. Glutamine, Serine, and Tyrosine can form hydrogen bonds due to their polar side chains. Tyrosine, with its aromatic ring, can also participate in hydrophobic interactions and pi-stacking. Proline, Isoleucine, Alanine, Phenylalanine, and Leucine primarily contribute through hydrophobic interactions. These residues tend to cluster together to exclude water, driving complex formation and contributing significantly to the stability of the interface. , The burying of hydrophobic surface area upon complex formation is a major driving force for protein association. The flexibility of amino acid residues at binding sites is also crucial for many protein functions. Conformational changes upon binding can optimize interactions and achieve biological function.

1.

Pymol interaction analysis of (a) chain D&F of 4XR8 pdb structure, Green: chain D, Blue: chain F, Red: binding sites. PRO 5, GLN 6, GLU 7, ARG 8, ARG 10, PRO 13, GLN 14, GLU 18, ILE 23, HIS 24, ARG 40, TYR 43, ASP 44, ALA 46, PHE 47, ASP 44, ALA 46, PHE 47, ASP 49, TYR 92, SER 97, ASP 98, LEU 100, PRO 109, LEU 110, SER 111,PRO 112,GLU 113, LYS 115 (b) A monomeric subunit. α–helices = 49, β–sheets = 22, coil/loop = 64.

3.2. Identification of Novel Sites on Chain F of 4XR8 PDB Structure

The potential binding pockets of the HPV16 E6 protein were quantitatively evaluated using DogSiteScorer. The analysis identified six potential cavities, with P_0 demonstrating the most favorable characteristics for aptamer binding. P_0 yielded a DrugScore of 0.79, a total volume of 578.11 ų, and a surface area of 744.31 Ų. These values were significantly higher than those of the subsequent pockets, such as P_1 (DrugScore: 0.29, Volume: 227.33 ų) and P_2 (DrugScore: 0.40, Volume: 141.57 ų), suggesting that P_0 provides the most suitable environment for high-affinity interactions.

Figure illustrates the predicted “novel sites” on Chain F of the 4XR8 protein, specifically highlighting Histidine residues at positions 78, 118, and 126. These predictions were generated computationally using Charm GUI web server tools, indicating the application of molecular modeling techniques to identify potentially significant regions within the protein structure. The designation of these sites as “novel” is particularly noteworthy. In structural biology, a novel site often implies a region of a protein not previously recognized for its functional importance or involvement in specific molecular interactions. The identification of such sites can open new avenues for understanding the biological role of a protein, its mechanism of action, or its potential as a target for therapeutic intervention. The specific identification of Histidine residues (HIS 78, HIS 118, HIS 126) at these novel locations is also significant. Histidine is a unique amino acid due to its imidazole side chain, which allows it to act as both a proton donor and acceptor. Its pK _a value often makes it sensitive to physiological pH changes. Consequently, Histidine residues are frequently found in critical functional regions of proteins, including enzyme active sites, metal-binding sites, and interfaces involved in proton transfer or allosteric regulation. Their presence at these predicted novel sites on Chain F suggests that these specific positions might play an as-yet-undiscovered role in the function, stability, or interaction profile of the 4XR8 protein. For example, they could be involved in transient interactions with other molecules, or they might contribute to the precise conformational changes necessary for the activity of the protein. The reliance on Charm GUI web server tools for this prediction underscores the growing importance of computational methods in modern structural biology research. Such tools, often based on sophisticated force fields and simulation algorithms, enable researchers to analyze complex protein structures, predict potential interaction sites, and generate testable hypotheses that guide experimental design. , While these predictions require experimental validation, they serve as a crucial first step in identifying promising areas for focused research.

2.

Chain F novel sites. HIS 78, HIS 118, HIS 126 predicted by Charm Gui.

3.2.1. Structural Validation of Chain F

The initial 3D model of the 4XR8 protein was subjected to rigorous stereochemical validation using the SAVES v6.1 server. Preliminary Ramachandran plot analysis indicated that only 87.1% of residues were in the most favored regions, with two residues (ARG 99 and ALA 41) located in disallowed regions. To resolve these local geometric distortions and ensure a high-fidelity structure for docking, the model was refined using the GalaxyRefine server, which utilizes side-chain repacking and mild molecular dynamics relaxation. , Postrefinement analysis revealed a significant improvement in the model’s structural integrity. The final Ramachandran plot (Figure c) showed that 94.2% (130/138 residues) of nonglycine and nonproline residues fall within the most favored regions, with 0.0% in disallowed regions. This exceeds the standard benchmark of >90% required for high-quality protein models. Further validation confirmed the superior reliability of the refined structure. The ERRAT overall quality factor improved to 97.8723, indicating a highly accurate distribution of nonbonded interactions. Additionally, the QMEAN score of 0.80 ± 0.007 reflects excellent structural fidelity and agreement with native-like protein features. These comprehensive validation results confirm that the refined 4XR8 structure is a robust and accurate foundation for the subsequent molecular docking and MD simulation studies.

3.

Structural Validation of the protein Model. (a) WHATCHECK Quality Check summary where each numbered box corresponds to a specific check. The color of each box indicates the quality rating for that check: Red (problematic area), Yellow (warning), and Green (good quality parameter within acceptable ranges). (b) Local Quality Profile: This bar graph illustrates the local quality of the protein backbone conformation across its sequence. The y-axis represents an “Error value” (likely related to unfavorable local structural parameters, such as bond lengths, bond angles, or dihedral angles) expressed as a percentage, while the x-axis indicates the residue number. Higher bars, especially those crossing the 95% threshold lines, suggest regions with poorer local geometry or potential structural issues. The color coding of the bars (red, yellow, gray) also likely correlates with the quality assessment (red for significant errors, yellow for warnings). (c) Ramachandran Plot: This plot shows the distribution of the backbone dihedral angles, phi and psi, for all amino acid residues in the protein model. Each black dot represents a single residue (d) A global quality estimate of 0.80 ± 0.007, indicating that the refined model possesses high structural fidelity and agreement with the expected features of native-like proteins.

3.3. Molecular Docking

3.3.1. Virtual Screening

The structural modeling and subsequent virtual screening were conducted for eight lead candidates. Apt4 was omitted from this phase due to the lack of a reliable 3D starting structure, resulting in the final selection of Apt1–Apt3 and Apt5–Apt9 for comprehensive MD evaluation. Figure depicts aptamer complexes with the 4XR8 protein (HPV16 E6), a direct and critical output of our virtual screening process. Our research employed virtual screening methods, including tools like AutoDock Vina, to initially screen a library of aptamers against the 4XR8 protein, identifying top-ranked candidates. The presented complexes are the refined results of further molecular docking, specifically using HADDOCK, for these selected aptamers. This entire computational workflow aims to minimize the laborious effort required for experimental screening by prioritizing the most promising aptamers. Each subimage visually represents a predicted binding site, illustrating how a particular aptamer candidate is hypothesized to interact with the E6 protein. These visualizations, generated by flexible docking approaches like HADDOCK, provide crucial insights into the precise orientation and contacts between the aptamer (orange stick models) and the protein (light blue ribbon). Understanding these binding modes is fundamental to rational aptamer design and validation.

4.

Predicted Binding Modes of aptamer candidates to HPV-16 E6 Oncoprotein (PDB: 4XR8). Light Blue-Ribbon Structure: Represents the HPV-16 E6 protein, which is part of the oncogenic E6/E6AP/p53 ternary complex. Orange Stick Models-DNA aptamer candidates, Amino Acid Labels (ARG-8, TYR-32, CYS-51)- Indicate specific residues on the protein identified as interacting with the aptamer. Blue Linesrepresent various noncovalent interactions and close contacts between the aptamer and the protein, contributing to the overall binding stability. Dashedellow linesYellow Lines - predicted hydrogen bond interactions between the aptamer and the protein, crucial for specific recognition and binding affinity.

The blue and yellow dashed lines, along with the labeled amino acid residues, are not just arbitrary lines; they signify specific molecular interactions (like hydrogen bonds, electrostatic interactions, and van der Waals forces) that contribute to the overall binding affinity and stability of the aptamer-protein complex. These detailed interaction maps are essential for interpreting the quantitative docking scores, which are numerical estimations of binding strength used to rank aptamer candidates in virtual screening. A more negative docking score, reflecting a more stable complex, aligns with the visual evidence of robust interactions. By visualizing aptamers binding to the 4XR8 structure, which represents E6 in its oncogenic complex with E6AP and p53, Figure provides critical insight into how these aptamers might disrupt the function of E6. The predicted binding sites and interaction types offer a hypothesis for how the aptamers could interfere with the E6-mediated degradation of p53, thereby informing their therapeutic potential against HPV16-associated cancers.

3.3.2. HADDOCK

Table details the specific amino acid residues of the HPV 16 E6 oncoprotein (4XR8) that are predicted to form key molecular interactions with various computationally designed aptamer candidates (Apt1, Apt2, Apt3, Apt5, Apt6, Apt7, Apt8, Apt9) and the positive control F2. These interactions were identified through molecular docking as part of a virtual screening strategy, which aims to minimize experimental effort and accelerate the discovery of aptamers. , The table categorizes interactions into specific hydrogen bonds formed between the aptamer and E6 residues, critical for molecular recognition and stability of ligand–receptor bonds. Electrostatic interactions between oppositely charged E6 residues (such as Arginine, Lysine) and the charged phosphate backbone of the aptamer contribute significantly to binding specificity and affinity. Interactions involving the electron-rich aromatic rings of certain E6 residues (for example, Tyrosine) and positively charged aptamer components or E6 residues (for example, Arginine). These play an important role in modulating molecular recognition and complex stabilization. Nonpolar contacts between hydrophobic E6 residues (for example, Proline, Isoleucine, Leucine) and the aptamer, contributing to the overall stability of the aptamer-protein interface by minimizing contact with water. The identification of these specific interaction types and involved residues provides crucial insights into the molecular basis of aptamer binding to E6, further validating the in-silico design process and informing the potential mechanisms by which these aptamers could interfere with the oncogenic function of E6.

1. Predicted Molecular Interactions between Selected Aptamer Candidates and HPV-16 E6 Oncoprotein (PDB: 4XR8)­ .

aptamers	H-bond interaction	salt bridge	π-cation	hydrophobic interaction
Apt1	TYR 32, CYS 51, 3(ARG) 55, 2 (SER) 74, HIS 78, GLN 107, ARG 129, ARG 131.	2 (ARG) 8, ARG 10, 2(ARG)129, ARG 131.	-	-
Apt2	3(ARG)8, 2 (ARG) 10, 2 (TYR)32, 2 (CYS) 51, 2 (ILE) 52, TYR 70, TYR 76, ARG 77, 2(HIS) 78, 2(ARG) 129, ARG 131.	ARG 102, ARG 131, ARG 146.	-	ARG 77
Apt3	PRO 5, 2(ARG) 10,TYR 32, ARG 55, 2(SER)74.	ARG 102, 2(ARG) 129, ARG 131.	-	-
Apt5	PRO 5, GLN 6, 4(ARG) 10, GLN 14, TYR 32, CYS 51, TYR 54,2(SER) 71, SER 74, TYR 92, ARG 129, GLY 130, ARG 131.	ARG 8, LYS 11, 2(ARG) 55, ARG 102.	ARG 10, ARG 131	-
Apt6	2(GLN) 6, 3(ARG) 8, 2(ARG) 10, TYR 32, ASP 49, TYR 54, ARG 55, 2(SER) 71, SER 74, HIS 78, GLN 107, 2(ARG) 129, ARG 131.	ARG 10, ARG 55, HIS 78, ARG 131.	-	-
Apt7	3(ARG) 10, TYR 32, CYS 51, 2(ARG) 55, 2(SER) 74, TYR 92, GLN 107, ARG 129, GLY 130, ARG 131	ARG 8, 2(ARG) 102, 2(ARG) 129, ARG 131.	-	ARG 131
Apt8	PRO 5, 2(ARG) 10, LYS 11, 2(TYR) 32, TYR 54, ARG 55, TYR 70, SER 74, HIS 78, ARG 102, 2(ARG) 129, ARG 131.	ARG 8, 2(ARG) 10, 2(ARG) 77, ARG 78, ARG 102, ARG 131.	-	VAL 53
Apt9	PRO 5, GLN 6, 4(ARG) 10, 2(GLN) 14, TYR 32, ASP 49, LEU 50, CYS 51, SER 74, HIS 78, LEU 100, ARG 102, LYS 115, ARG 131.	2(ARG) 8, ARG 10, LYS 11, ARG 102.	2 (ARG) 131	-
F2	2(ARG) 8, ARG 10, LYS 11, 2(GLN) 14, GLU 18, ARG 40, TYR 43, ILE 101, 2 (SER) 111	ARG 10, 2(LYS) 11, ARG 102, LYS 108, LYS 115	2(TYR) 43, PHE 47	PHE 47

^a= Not available.

3.4. MD Simulation of Aptamer Complexes

3.4.1. Root Mean Square Deviation (RMSD) Analysis of Aptamer-Protein Complexes

Figure (a) illustrates the Root Mean Square Deviation for the entire aptamer-protein complexes over 250 (ns) of MD simulation. The RMSD, a measure of the average distance between a set of atoms relative to a reference structure, is presented in nanometers (nm) on the y-axis. In this case, the RMSD is calculated for the entire complex, with the initial protein structure serving as the reference. This provides critical insights into the structural stability and conformational changes of both the aptamer and the protein upon complex formation. Typically, MD simulations show an initial equilibration phase where the system adjusts from its starting configuration, resulting in an initial rise in RMSD. Following this, a stable plateau indicates that the system has reached equilibrium, meaning its structural fluctuations are now centered around an average conformation. All the simulated complexes show this initial rise, mostly within the first 20–50 ns, after which they generally stabilize, albeit with varying degrees of fluctuations. −

5.

Molecular Dynamics analysis of HPV16 E6-aptamer complexes.(a) Structural stability showing Apt 8 and Apt 9 (0.23 nm) significantly outperformed the F2 control (0.94 nm), (b) Local residue flexibility of the 4XR8 oncoprotein, (c) Global compactness of the complexes (d) Interface persistence; Apt 8 (11.49) and Apt 9 (9.66) exceeded the F2 benchmark (8.59). (e) SASA: Solvent exposure; Apt 9 maintained the most buried interface (99.95 nm²). Note: Apt4 was excluded from the study at the structural modeling stage and is therefore not represented in the dynamic analysis.

The salmon pink line, representing the protein alone, shows a relatively stable RMSD profile, fluctuating around 0.15 nm after the initial equilibration phase. This indicates that the protein maintains its structural integrity and remains rigid when considered as part of the complex. In contrast, the F2 positive control exhibited significant conformational instability, with RMSD values fluctuating between 0.8 and 1.2 nm and averaging 0.94 ± 0.19 nm. Candidates Apt7, Apt3, and Apt8 demonstrated the most stable RMSD profiles. Apt7 consistently hovered below 0.2 nm, indicating a highly stable binding interaction with minimal deviation from the reference structure. Apt3 and Apt8 also showed robust stability, with values typically ranging between 0.15 and 0.25 nm. These consistent plateaus suggest well-defined, rigid interactions, which are desirable characteristics for therapeutic or diagnostic applications. Conversely, Apt1, Apt9, Apt2, and Apt5 displayed intermediate to high flexibility, with fluctuations ranging from 0.25 nm to over 0.4 nm. Apt6 exhibited the highest instability among all complexes, frequently exceeding 0.5 nm and reaching 0.7 nm. This excessive flexibility suggests unfavorable accommodation within the protein binding pocket or a potentially transient interaction, identifying Apt6 as a candidate that may require optimization.

3.4.2. Root Mean Square Fluctuation (RMSF) Analysis of Aptamer-Protein Complexes

The RMSF profiles illustrate varying degrees of flexibility across the complexes (Figure b). Generally, all complexes exhibited higher fluctuations at the N-terminal and C-terminal regions, a common observation reflecting the relative lack of constraints at protein termini. On the other hand, many complexes showed lower RMSF values in the central regions (residues 60–120), suggesting a more rigid core structure. The intrinsic flexibility of the protein backbone appeared to be modified by aptamer binding, as its fluctuations remained within a moderate range throughout the trajectory. The Apt6 complex consistently showed the highest flexibility, with pronounced peaks at residues 20, 40, and 100, and a significant fluctuation exceeding 0.7 nm at residue 145. Similarly, Apt2 exhibited localized instability at the C-terminus with a high peak around residue 150. This widespread local fluctuation aligns with their high global RMSD, indicating that these complexes undergo significant structural rearrangements or possess a less stable binding interaction. Apt5 also demonstrated noticeable fluctuations, particularly in the N-terminal region (residues 10–20), reflecting intermediate flexibility compared to the more stable candidates. In contrast, Apt7 displayed one of the lowest and most stable RMSF profiles, with fluctuations in the central regions rarely exceeding 0.2 nm. Candidates Apt3 and Apt8 followed a similar trend of contained fluctuations, while Apt1 and Apt9 showed slightly more mobility toward the termini. These consistently low RMSF values across the structure, particularly at potential protein-aptamer interface regions, signify highly rigid and stable complex formations. This analysis complements the RMSD data by identifying the specific residues contributing to the stability of the lead candidates.

3.4.3. Radius of Gyration (R _g) Analysis of Aptamer-Protein Complexes

The radius of Gyration measures the effective size and compactness of a molecule or complex. A smaller R _g value indicates a more compact structure, while a larger R _g suggests a more extended or unfolded conformation. Changes in R _g over time can signal structural transitions or alterations in the overall shape and density of the complex. As shown in Figure c, the protein maintained the lowest and most stable R _g profile, fluctuating consistently around 1.45 nm, which provides a baseline for evaluating the impact of aptamer binding. Candidates Apt7 and the F2 control exhibited the highest R _g values, ranging between 1.52 and 1.65 nm, with occasional spikes reaching 1.80 nm. Interestingly, while Apt7 demonstrated high stability in RMSD and RMSF metrics, its elevated R _g suggests it adopts a stable yet “open” or extended conformation upon binding. Conversely, the Apt6 complex showed an intermediate but highly dynamic R _g profile (1.45–1.55 nm). These fluctuations, alongside its high RMSD, reinforce the conclusion that Apt6 undergoes significant structural transitions and lacks a tightly constrained binding mode. The remaining candidates, Apt8, Apt3, Apt1, and Apt9, formed relatively compact and stable complexes. Apt3 and Apt8 maintained the most tightly packed structures with R _g values typically below 1.50 nm, closely mirroring the protein’s own compactness. Apt1 and Apt9 showed slightly higher values (1.50–1.53 nm), while Apt2 and Apt5 displayed moderate fluctuations reaching 1.55 nm, indicating greater conformational adaptability.

3.4.4. Hydrogen Bond Analysis of Aptamer-Protein Complexes

Figure (d) presents the number of intermolecular hydrogen bonds formed between each aptamer and the protein over 250000 ps of MD simulation. The y-axis indicates the number of hydrogen bonds, while the x-axis represents simulation time in (ps). Hydrogen bonds are crucial noncovalent interactions that significantly contribute to the stability and specificity of protein–ligand complexes. A higher and more consistent number of hydrogen bonds generally implies a stronger and more stable binding interaction, while fluctuations can indicate dynamic rearrangements at the binding interface. All aptamer-protein complexes exhibit fluctuations in the number of hydrogen bonds over time, a characteristic of dynamic systems. However, there are clear differences in the average number of hydrogen bonds maintained and the extent of these fluctuations among the different aptamers

Candidates Apt3 and Apt8 consistently maintained the highest density of interactions, typically fluctuating between 10 and 20 bonds with frequent peaks exceeding 15. These results significantly outperformed the F2 positive control, which averaged 8.59 ± 1.94 bonds. Conversely, Apt1, Apt2, Apt5, and Apt9 showed intermediate bonding profiles (6–16 bonds), while Apt6 exhibited the weakest network, with counts frequently dropping below 5 and occasionally to zero. This lack of sustained hydrogen bonding for Apt6 aligns with its high flexibility and instability observed in the previous metrics. The case of Apt7 is particularly noteworthy; despite demonstrating high conformational stability (low RMSD/RMSF), it maintained a very low number of direct hydrogen bonds. Coupled with its high R _g and SASA, this suggests that Apt7 forms a stable but “open” complex that likely relies on alternative stabilization mechanisms, such as extensive hydrophobic contacts or water-mediated networks, rather than a dense direct hydrogen-bonding interface. Overall, the robust hydrogen bonding in Apt3 and Apt8 reinforces their status as the most promising lead candidates for high-affinity binding.

3.4.5. Solvent Accessible Surface Area (SASA) Analysis of Aptamer-Protein Complexes

SASA measures the surface area of a molecule that is accessible to a solvent. It provides insights into the compactness of the system and how much of its surface is exposed to the aqueous environment. Changes in SASA can reflect structural rearrangements that lead to burying or exposing parts of the molecule, and in a complex, a decrease in SASA often indicates the formation of a binding interface where surfaces become hidden. −

As shown in Figure e, most systems stabilized after the initial 50 ns equilibration. The protein baseline was established at 94–98 nm², with all aptamer-protein complexes exhibiting higher values due to the additional surface area provided by the aptamer sequences. Apt7 consistently exhibited the highest SASA values, fluctuating between 104 and 122 nm². This high exposure strongly correlates with its elevated Radius of Gyration, confirming that Apt7 adopts a stable but unusually extended or ’open’ conformation. In contrast, the Apt6, Apt2, and Apt5 complexes displayed moderately high and fluctuating profiles (100–112 nm²). These dynamic fluctuations, particularly the increase in Apt2 SASA between 80–100 ns, align with their high RMSD and RMSF, reflecting a less compact and more flexible binding mode. Conversely, Apt3 and Apt8 maintained the lowest SASA profiles (96–104 nm²). This minimal solvent exposure, combined with their low RMSD and R _g, signifies the formation of highly compact and well-packed complexes where substantial parts of the binding interface are buried. Apt1 and Apt9 also demonstrated relatively stable profiles within this lower range. Collectively, these metrics identify Apt3 and Apt8 as forming the most rigid and tightly integrated structures, while Apt7 remains unique for its stable but high-exposure binding mechanism.

3.4.6. Statistical Analysis and Stability of 4XR8-Aptamer Complexes

The dynamic behavior of the HPV16 4XR8-aptamer complexes was evaluated to be over 250 ns in Table to ensure the systems reached a stable equilibrium ensemble. To address the requirement for statistical rigor, we report the Mean ± Standard Deviation and 95% Confidence Intervals for all primary trajectories. The convergence of these parameters across the extended simulation time frame provides a statistically sound foundation for comparing the binding efficacy of the candidate aptamers. The RMSD analysis revealed that most of the 4XR8-aptamer complexes, particularly Apt2, Apt8, and Apt9, exhibited superior structural stability compared to the apoprotein (0.39 ± 0.07 nm). Notably, Apt9 maintained a highly consistent backbone geometry with a mean RMSD of 0.23 ± 0.03 nm and a narrow 95% CI (0.23–0.23 nm), signifying that the system remained within a single, well-defined energetic minimum.

2. Quantitative Summary of the Molecular Dynamics Parameters for the HPV16 E6-Aptamer Complexes, the F2 Positive Control, and the Apo-Protein .

	RMSD		RMSF		R g		HB		SASA
MD	Mean ± SD	95% CI	Mean ± SD	95% CI	Mean ± SD	95% CI	Mean ± SD	95% CI	Mean ± SD	95% CI
Apt1	0.24 ± 0.03	0.22–0.24	0.14 ± 0.0666	0.13–0.15	1.52 ± 0.01	1.52–1.52	7.36 ± 1.93837	7.28–7.44	100.22 ± 1.77	100.16–100.29
Apt2	0.23 ± 0.06	0.22–0.23	0.15 ± 0.11	0.13–0.17	1.53 ± 0.02	1.52–1.53	10.13 ± 2.46	9.98–10.28	100.96 ± 3.80	100.72–101.19
Apt3	0.27 ± 0.05	0.26–0.27	0.16 ± 0.06	0.15–0.17	1.51 ± 0.02	1.51–1.51	10.27 ± 3.13	10.12–10.42	99.49 ± 1.82	99.41–99.58
Apt5	0.27 ± 0.08	0.27–0.28	0.22 ± 0.07	0.21–0.23	1.53 ± 0.03	1.52–1.53	10.55 ± 2.09	10.46–10.63	103.12 ± 3.44	102.98–103.27
Apt6	0.31 ± 0.07	0.30–0.31	0.15 ± 0.05	0.15–0.16	1.47 ± 0.03	1.47–1.47	8.39 ± 2.09	8.29–8.48	101.03 ± 2.27	100.93–101.13
Apt7	0.52 ± 0.10	0.51–0.52	0.35 ± 0.19	0.32–0.38	1.65 ± 0.05	1.64–1.65	8.25 ± 2.39	8.18–8.33	110.11 ± 3.56	109.99–110.22
Apt8	0.23 ± 0.06	0.22–0.23	0.16 ± 0.07	0.15–0.17	1.46 ± 0.02	1.46–1.46	11.49 ± 2.35	11.42–11.56	96.41 ± 2.05	96.35–96.47
Apt9	0.23 ± 0.03	0.23 −0.23	0.17 ± 0.08	0.16–0.18	1.51 ± 0.02	1.50–1.51	9.66 ± 2.5	9.57–9.75	99.95 ± 2.207	99.86–100.03
F2	0.94 ± 0.19	0.92–0.94	0.21 ± 0.09	0.16–0.18	1.46 ± 0.09	1.51–1.52	8.59 ± 1.94	8.52–8.67	115.13 ± 4.46	96.24–115.30
Prot	0.39 ± 0.07	0.38–0.39	0.17 ± 0.07	0.20–0.23	1.52 ± 0.01	1.46–1.46	-	-	96.32 ± 1.84	114.95–96.39

^aValues represent the mean ± standard deviation (Mean ± SD) and the 95% confidence interval derived from three independent replicates (n = 3). - not available.

While the positive control, F2, demonstrated a higher mean RMSD (0.94 ± 0.19 nm), its 95% CI (0.92–0.94 nm) indicates that this increased flexibility was sampled uniformly across the trajectory without causing structural disintegration. This suggests that while F2 allows for broader conformational sampling, the candidate aptamers like Apt8 (0.23 ± 0.06 nm) induce a more rigid, stabilized state in the E6 protein, which is often a prerequisite for effective inhibition of oncogenic interactions. The R _g and SASA metrics were utilized to assess the impact of aptamer binding on protein folding and surface exposure. The protein in the Apt8 complex achieved the highest degree of compaction (1.46 ± 0.02 nm), which was significantly lower than the values observed for Apt1 and Apt2. The stability of the R _g values, supported by SDs as low as ± 0.01, confirms that the protein maintained its globular fold throughout the 250 ns simulation. Furthermore, the binding of Apt8 resulted in a substantial reduction of the protein’s SASA to 96.41 ± 2.05 nm², compared to the apoprotein (115.13 ± 1.84 nm²). This statistical reduction in solvent exposure suggests that Apt8 effectively occupies a large portion of the E6 surface, potentially shielding critical binding motifs from their natural cellular targets. HB analysis provides a direct measure of the interaction strength between the aptamers and the 4XR8. Apt8 demonstrated the most robust interaction profile, maintaining an average of 11.49 ± 2.35 hydrogen bonds with a 95% CI of 11.42–11.56. This outperforms both the F2 positive control (8.59 ± 1.94) and other candidates such as Apt1 (7.36 ± 1.94), suggesting that Apt8 forms a more persistent and chemically favorable interface.

3.5. Molecular Mechanics Poisson–Boltzmann Surface (MMPBSA)

The Molecular Mechanics Poisson–Boltzmann Surface Area method provides a robust framework for estimating protein–ligand binding affinities by decomposing the total binding free energy into various physically meaningful components. Analyzing these terms allows for a detailed understanding of the energetic landscape governing aptamer-protein recognition. ,

3.5.1. Total Binding Free Energy

The calculated binding free energy (G _bind) values serve as a quantitative estimate of the overall binding affinity, where more negative values represent stronger and more favorable molecular interactions. In this study, Apt9 emerged as the primary lead candidate, exhibiting the most favorable binding free energy of −143.76 kcal/mol. This was followed by a cluster of high-affinity sequences that demonstrated significant energetic stability, including Apt7 (−138.71 kcal/mol), Apt1 (−138.26 kcal/mol), Apt5 (−137.76 kcal/mol), and Apt8 (−136.43 kcal/mol). The high absolute magnitude of these values suggests that these aptamers form energetically preferred complexes with the E6 oncoprotein, likely driven by optimized van der Waals and electrostatic contributions within the binding pocket. In contrast, a secondary group comprising Apt2 (−110.56 kcal/mol) and Apt6 (−108.36 kcal/mol) showed comparatively lower affinity, while Apt3 yielded the least favorable binding free energy at −99.97 kcal/mol. The relatively lower energetic magnitude for Apt3 indicates it forms the weakest complex among the candidates, potentially due to less optimal spatial orientation or weaker intermolecular contacts. These variations highlight the critical influence of specific nucleotide interactions and solvation effects in determining the ultimate binding strength to the E6 target. To contextualize these findings, the candidate aptamers were benchmarked against the literature-reported F2 RNA aptamer, which served as a positive control. The MM/PBSA analysis confirmed that Apt9 (−143.76 kcal/mol) successfully surpassed the thermodynamic affinity of the F2 control (−142.80 kcal/mol). Beyond its superior binding energy, Apt9 also demonstrated exceptional structural consistency throughout the 250 ns Molecular Dynamics trajectory, maintaining a low and stable RMSD of 0.23 nm and a robust hydrogen bonding network. This synergy between superior thermodynamic affinity and high conformational rigidity distinguishes Apt9 as a potent novel lead, offering a distinct competitive advantage over established benchmarks for the potential inhibition of the HPV16 E6 oncoprotein.

3.5.2. van der Waals Energy

The term E _vdw (Table ) reflects the steric complementarity and hydrophobic interactions at the binding interface. Favorable (negative) E _vdw values indicate good physical fit and close packing. Apt5 (−89.31 kcal/mol), Apt8 (−81.59 kcal/mol), and Apt1 (−77.40 kcal/mol) exhibit the most favorable E _vdw. This suggests that these aptamers form well-packed interfaces with the protein, maximizing contact surface area and nonpolar interactions, which are crucial for stable complex formation. Apt3 (−45.93 kcal/mol) and Apt7 (−59.38 kcal/mol) show less favorable E _vdw. For Apt3, despite its previously noted stability, this term is noticeably less negative, implying that its binding mechanism might rely more heavily on other forces rather than extensive surface burial. This contrasts with Apt5, which appears to leverage both strong van der Waals and electrostatic forces.

3.5.3. Electrostatic Energy

The E _elec term quantifies the direct electrostatic attraction or repulsion between the charged groups of the aptamer and protein in a vacuum. Given the polyanionic nature of aptamers, highly favorable (large negative) E _elec is expected and plays a primary role in initial recognition and binding. All aptamers show very large negative contributions, ranging from −3035.04 kcal/mol to −3513.40 kcal/mol. This confirms that electrostatic interactions are a major driving force for the association of these nucleic acid aptamers with the protein. Apt5 (−3513.40 kcal/mol), Apt1 (−3486.85 kcal/mol), and Apt7 (−3470.27 kcal/mol) exhibit the most negative E _elec. This strong gas-phase electrostatic attraction suggests a high degree of charge complementarity between these aptamers and the protein binding site. Apt6 (−3035.04 kcal/mol) has the least negative E _elec, aligning with its overall weaker binding.

3.5.4. Polar Solvation Energy

The ΔEPB term represents the energy cost associated with desolvating the charged and polar groups of the aptamer and protein upon complex formation. This term is typically unfavorable (positive) and often largely cancels out the favorable gas-phase electrostatic interactions. All aptamers incur a large positive (unfavorable) ΔEPB, ranging from 3039.50 to 3523.73 kcal/mol. This highlights the substantial energy penalty for removing water molecules from the highly charged surfaces of both the aptamer and the protein as they come together. Noticeably, aptamers with very favorable E _elec (Apt1, Apt5, Apt7) also tend to have very large positive ΔEPB values that are nearly equal in magnitude to their ΔEPB. This near cancellation is a well-known phenomenon in aqueous systems, where the favorable electrostatic interactions in a vacuum are largely compensated by the unfavorable desolvation energy. For Apt3 (3325.65 kcal/mol) and Apt8 (3356.66 kcal/mol), their ΔEPB values are comparable to their E _elec, reinforcing this cancellation effect.

3.5.5. Nonpolar Solvation Energy

The term ΔENPOLAR accounts for the hydrophobic effect and the reduction of solvent-accessible nonpolar surface area upon binding. This term is generally favorable (negative) because the burial of nonpolar surface area releases ordered water molecules back into the bulk solvent. All aptamers showed a favorable (negative) ΔENPOLAR contribution. This ranges from −34.74 kcal/mol (for Apt3) to −58.78 kcal/mol (for Apt5). The favorable nature of the nonpolar solvation terms indicates that the hydrophobic effect plays a consistent and important role in stabilizing the binding of all aptamers to the protein. This term provides a net positive contribution to the overall binding free energy, promoting complex formation. For instance, the strong, favorable nonpolar solvation term (−55.50 kcal/mol) of Apt8, combined with its other strong interactions, is a key factor in (Table ).

3. Energy Components and Calculated Binding Free Energies of Aptamer-Protein Complexes .

candidate aptamers	E vW (kcal/mol)	E Elec (kcal/mol)	ΔE PB (kcal/mol)	ΔE NPOLAR (kcal/mol)	calculated ΔG bind (kcal/mol)
Apt1	–77.40 ± 4.13	–3486.85 ± 74.61	3480.17 ± 67.93	–54.18 ± 2.90	–138.26
Apt2	–66.75 ± 11.38	–3147.21 ± 96.86	3147.18 ± 94.25	–43.78 ± 6.54	–110.56
Apt3	–45.93 ± 13.15	–3344.95 ± 43.01	3325.65 ± 130.30	–34.74 ± 9.30	–99.97
Apt5	–89.31 ± 12.96	–3513.40 ± 86.71	3523.73 ± 85.39	–58.78 ± 9.03	–137.76
Apt6	–68.42 ± 13.08	–3035.04 ± 159.51	3039.50 ± 160.20	–44.40 ± 8.41	–108.36
Apt7	–59.38 ± 20.52	–3470.27 ± 153.60	3435.45 ± 163.44	–44.51 ± 9.91	–138.71
Apt8	–81.59 ± 7.81	–3356 ± 98.50	3356.66 ± 99.48	–55.50 ± 6.55	–136.43
Apt9	–74.58 ± 7.04	–3117.29 ± 85.70	3102.19 ± 87.29	–54.08 ± 3.70	–143.76
F2	–86.48 ± 4.75	–2127.76 ± 126.53	2128.22 ± 117.29	–56.78 ± 3.83	–142.80

^aThe standard error (±) is reported for each energy component based on the variance observed across the trajectory frames used for the calculation. The ΔG _bind is calculated according to the thermodynamic cycle summation of the individual terms E _vW+ E _Elec+ ΔE _PB + ΔE _ENPOLA.

3.6. Principal Component Analysis of Aptamer Conformational Dynamics

Principal Component Analysis was performed on the MD simulation trajectories of 9 aptamer candidates (F2, Apt1-Apt9) to elucidate their major conformational motions. This method is widely employed in the analysis of biomolecular simulations to reduce the dimensionality of complex data and identify the most important collective movements in a system. PCA transforms high-dimensional data into a new set of orthogonal variables, called principal components, that capture the most significant variance in the dynamics of the system. By projecting the high-dimensional conformational space onto a few principal components, we can gain insights into the flexibility, stability, and conformational preferences of molecules. The 2D projection shown in Figure utilizes the first two principal components, PC1 (projection on eigenvector 1) and PC2 (projection on eigenvector 2), which collectively represent the largest collective motions of the aptamers during the simulations. The axes are given in nanometers (nm), indicating the magnitude of displacement along these principal modes. Each point in the plot corresponds to a single simulation snapshot (or frame), with each aptamer distinguished by a unique color and marker, as detailed in the legend.

6.

Principal Component Analysis of Conformational Dynamics for 9 Aptamer Candidates (Apt1-Apt9) during MD Simulations. This two-dimensional (2D) projection shows the conformational landscape explored by 9 distinct DNA aptamer candidates during MD simulations. The x-axis represents the projection onto eigenvector 1 and the y-axis represents the projection onto eigenvector 2, both in nanometers (nm). PC1 and PC2 capture the largest amplitude collective motions of the aptamers. Each point on the plot corresponds to a single simulation snapshot, with individual aptamers differentiated by distinct colors as indicated: Apt1 (blue), Apt2 (orange), Apt3 (green), Apt5 (red), Apt6 (violet), Apt7 (brown), Apt8 (purple), and Apt9 (gray). The clustering and distribution of points for each aptamer reflect their conformational stability and flexibility. Tightly grouped clusters (Apt6, Apt8) suggest higher structural stability and limited conformational sampling, while broadly dispersed distributions (Apt5, Apt7) indicate greater flexibility and exploration of a broader range of conformational states. F2 (black) occupies a large region.

The distribution of clusters in the PCA plot (Figure ) reveals the conformational landscape and collective motions of each aptamer-protein complex. The spatial spread of these clusters serves as a direct indicator of flexibility; a localized distribution signifies a rigid complex, while a wider spread represents increased conformational sampling. Apt8 exhibited the most concentrated cluster, aligning with its low RMSD of 0.23 nm, whereas the F2 control occupied a significantly larger region, confirming its more flexible nature. Apt6 and Apt8 maintained tight, localized clusters, suggesting a more rigid structural core with limited exploration of alternative conformations. In contrast, Apt7 (brown) and Apt5 (red) displayed broad, dispersed distributions across the lower-right and upper-right quadrants, respectively. This signifies that these candidates are highly flexible in this coordinate space, sampling multiple distinct structural states. Apt2 appeared highly restricted, suggesting minimal sampling, while Apt1, Apt3, and Apt9 exhibited moderate distributions indicative of a few major conformational substates. This PCA mapping provides a comprehensive overview of the dynamics of the complexes, distinguishing the candidates by their structural stability and the range of conformational space they explore during the simulation.

4. Conclusion

This research established a streamlined computational pipeline for selecting DNA aptamers against the HPV16 E6 oncoprotein, integrating virtual screening with 250 ns Molecular Dynamics simulations. By evaluating eight candidates against the F2 RNA benchmark, we identified Apt9 as the most potent lead, achieving a binding free energy of −143.76 kcal/mol and surpassing the −142.80 kcal/mol recorded for the positive control. Structural analyses prioritized Apt9 and Apt8 as the most stable complexes, both maintaining a minimal RMSD of 0.23 nm. Apt 8 further distinguished itself by maintaining the most robust hydrogen-bonding network (11.49 bonds), significantly outperforming the F2 control. Mechanistically, these aptamers target critical residues on the 4XR8 protein model, specifically Histidine at positions 78, 118, and 126 of Chain F. While protein model validation through PROCHECK and ERRAT confirmed a reliable foundation, the identified binding site specificity provides a clear roadmap for targeted therapy. Overall, this in silico approach designates Apt9 and Apt8 as high-priority candidates. Future experimental validation using Surface Plasmon Resonance and isothermal titration calorimetry will confirm these findings, advancing the development of precise diagnostic and therapeutic tools for cervical cancer management.

4.1. Limitations and Future Recommendations

The primary limitation of this study is its purely in silico nature, meaning the high-affinity interactions predicted for top candidates such as Apt8 and Apt9 have not yet been experimentally validated. To address these gaps, the top binding candidate aptamers should be validated using SPR, ITC, or MST. Ultimately, the candidate aptamers should be applied in the development of a rapid multiplexed aptasensor for E6 detection and explored as Aptamer-Drug Conjugates for targeted delivery to HPV16-positive cells.

The data presented in this manuscript can be requested from the corresponding authors on reasonable requests.

This research was funded by NRF South Africa and the UWC Nanotechnology platform.

The authors declare no competing financial interest.