Background

Single-walled carbon nanotubes (SWCNTs) are widely used in nanomaterials science due to their unique structural and physicochemical properties. However, their potential interactions with biomolecules and resulting biological effects remain insufficiently understood. This study aims to elucidate the interaction mechanisms between SWCNTs and model biomolecules using a combination of computational simulations and experimental approaches, providing insights into nanotoxicology and bio-nano interactions. Explore CD ComputaBio's integrated computational and experimental services for studying nanomaterial-protein interactions, offering innovative solutions to elucidate binding mechanisms, evaluate biocompatibility, and support the rational design of safer nano-enabled systems.

Computational Methods

Molecular Dynamics (MD) Simulation

MD simulations were performed using the AMBER simulation package to investigate the dynamic behavior and stability of the SWCNT-biomolecule complex. Prior to simulation, the system was prepared by removing crystallographic water molecules and adding missing hydrogen atoms using the tleap module. A model of the single-walled carbon nanotube was constructed and geometry-optimized using Visual Molecular Dynamics (VMD) software. The protein was described using the ff03 force field, while parameters for the SWCNT were assigned using the General Amber Force Field (GAFF). After initial energy minimization, the system underwent a stepwise equilibration protocol: heating from 0 to 300 K over 50 ps under NVT conditions, followed by 100 ps of density equilibration under NPT conditions. Finally, a production MD run of 100 ns was carried out under periodic boundary conditions in the NPT ensemble (1 atm, 300 K), employing a 2-fs time step, the SHAKE algorithm for hydrogen bonds, and the Particle Mesh Ewald (PME) method for long-range electrostatics. Trajectory snapshots were saved every 10 ps for subsequent analysis of stability, interactions, and dynamics.

MM/GBSA Binding Free Energy Calculation

To quantitatively evaluate the affinity and driving forces of the SWCNT-biomolecule interaction, the Molecular Mechanics/Generalized Born Surface Area (MM/GBSA) method was applied to frames extracted from the equilibrated portion (e.g., 60–100 ns) of the MD trajectory. Binding free energies between protein and ligand were computed using the MM/GBSA method, according to: ΔG_bind = ΔG_complex − ( ΔG_receptor + ΔG_ligand ) = ΔE_internal + ΔE_vdW + ΔE_elec + ΔG_GB + ΔG_SA. The nonpolar solvation energy was estimated from the solvent-accessible surface area (SASA). Entropic contributions were omitted due to high computational cost and associated uncertainty.

Binding Mode Analysis

Following MD simulation, the binding interface between SWCNT and the protein was analyzed in detail. A key metric, the contact surface area (S), was calculated to quantify the extent of the interaction interface. This area is defined as half the difference between the solvent-accessible surface area (SASA) of the SWCNT-protein complex and the sum of the SASAs of the isolated protein and SWCNT: S = ½ [ SASA( complex) - ( SASA( protein) + SASA(SWCNT))] . This value, monitored throughout the trajectory, provides a quantitative measure of the buried surface area upon binding, indicating the stability and intimacy of the nano-bio interface.

Residue Interaction Network (RIN) Analysis

To visualize and quantify changes in the protein's internal interaction network induced by SWCNT binding, Residue Interaction Network (RIN) analysis was performed using Cytoscape software alongside custom analytical scripts. Networks were constructed where nodes represent amino acid residues and edges represent non-covalent interactions (e.g., hydrogen bonds, van der Waals contacts, π-π stacking). For the apo (protein alone) and holo (protein-SWCNT complex) systems, interaction networks were generated from the simulated trajectories. Topological analysis, including changes in node degree (number of connections per residue) and overall network density, was conducted to assess how SWCNT binding perturbs the local and global interaction landscape of the protein, potentially leading to structural destabilization or functional modulation.

Computational Results Analysis

MD Simulation: Stability and Interaction Analysis

RMSD and radius of gyration analysis confirmed system stability during the 100 ns simulation. Following the molecular docking of SWCNT with BSA, 100 ns molecular dynamics (MD) simulations were performed for both the BSA-SWCNT complex and the APO-BSA system using the AMBER14 software package. To evaluate the overall stability of the complex systems, the Root Mean Square Deviation (RMSD) was calculated based on the data extracted from the simulation trajectories. As illustrated in Figure 1, both systems reached equilibrium and exhibited no significant fluctuations, indicating that the simulations achieved a stable state.

Figure 1. RMSD analysis between SWCNT and BSA.

The contact surface area between SWCNT and biomolecule remained consistent, indicating stable and persistent binding throughout the simulation. The contact surface area is denoted as S, which is defined as half the difference between the solvent-accessible surface area (SASA) of the SWCNT-BSA complex and the sum of the SASAs of BSA and SWCNT. From 0 ns to 40 ns, as the interface fluctuated, it was observed that SWCNT and BSA continuously adjusted their relative positions to achieve a more favorable binding mode. At 40 ns, S began to show a rapid increasing trend, indicating that SWCNT was progressively seeking a more stable binding site. Furthermore, from t = 40 ns to 100 ns, S exhibited only minor fluctuations around approximately 330 Ų. After 60 ns, the curve essentially plateaued, demonstrating that the contact surface of this domain had reached a relatively stable state. This suggests that SWCNT ultimately successfully inserted into the binding cavity of BSA and remained stably bound.

Figure 2. Binding mode of BSA-SWCNT. (The change of BSA-SWCNT contact surface area with time is shown in blue for BSA and red for SWCNT).

Binding Site Characterization

At binding site III, one tryptophan (Trp), seven tyrosines (Tyr), and four phenylalanines (Phe) are present. The single-walled carbon nanotube (SWCNT) is positioned within the hydrophobic cavity and is surrounded by a cluster of hydrophobic amino acids. These hydrophobic residues include Tyr137, Tyr139, Tyr160, Trp134, Phe126, Phe133, Phe164, Leu115, Leu122, Leu138, Leu153, Arg143, Arg144, and Ile141, as illustrated in Figure 3.

Figure 3. Binding mode of SWCNT in subdomain IB of BSA. The BSA structure is shown in ribbon representation.

Several amino acids were observed in close proximity to the SWCNT. Using CHIMERA software, the distances between the SWCNT and these residues were defined as the distance between their respective centers of mass (COM). As illustrated in Figure 4, Tyr137, Tyr160, Tyr139, and Trp134 are situated very close to the SWCNT, with measured distances of 6.506 Å, 6.954 Å, 9.141 Å, and 8.949 Å, respectively. Notably, Tyr137 and Tyr160 are initially buried within the hydrophobic cavity of BSA; however, the binding of the SWCNT progressively pushes them toward the surface, exposing them to a more polar environment. This observation is highly consistent with the synchronous fluorescence results, which indicated an increase in polarity and a decrease in hydrophobicity around the tyrosine residues. These findings provide a robust structural basis for explaining the fluorescence quenching of BSA in the presence of SWCNTs. This is in excellent agreement with the experimental data mentioned above, further confirming that hydrophobic interaction is the primary driving force governing the binding behavior between SWCNTs and BSA.

Figure 4. Distance plots between key amino acid residues and the SWCNT. The SWCNT is shown in yellow, and the amino acid residues are shown in blue.

Residue Interaction Network Analysis

To further analyze the interaction mode between amino acids near the binding cavity, we used the rinalyzer plug-in in Cytoscape software, which can provide us with a visual analysis of the amino acid network. The software can also be combined with chimera software for analysis, which is based on the network analysis of amino acid residues. This is a novel method to analyze the stability of the binding cavity. Residue-residue interactions are represented as edges in the network. The main interaction types include van der Waals force, hydrogen bond, ionic interaction, π-π stacking, salt bridge, etc. The middle node represents a single amino acid. Taking the centroid of single-walled nanotubes as the center, all amino acids within 5 Å around the centroid were selected, with a total of 28 amino acids. Network analysis was conducted on the binding site region (Site III/subdomain IB) before and after SWCNT binding. In the apo-BSA state, the interaction network within this pocket was dense and stable, featuring a total of 50 interaction edges (comprising hydrogen bonds, van der Waals contacts, and π-π stacking). Following SWCNT binding and 100 ns of MD simulation, the network was significantly perturbed. The total number of edges decreased to 31, representing a 38% reduction in intramolecular interactions within the binding site.

Figure 5. Residue interaction network (RIN) analysis of BSA-SWCNT. (A) Amino acid interaction network of the APO-BSA system. (B) SWCNT-BSA system, where the BSA structure is shown in blue and the SWCNT is shown in gold. Red lines represent hydrogen bonds, blue lines represent van der Waals forces, green lines represent ionic interactions, and black lines represent π-π stacking.

As shown in Figure 5, for the SWCNT-BSA system, among the 28 selected amino acids, there are a total of 31 edges, meaning 31 amino acid–amino acid interactions. These include 15 van der Waals interactions, 13 hydrogen bonds, and 3 ionic interactions. In contrast, in the system without single-walled carbon nanotubes (Apo-BSA), there are 50 edges, representing 50 amino acid–amino acid interactions, comprising 25 van der Waals interactions, 20 hydrogen bonds, and 5 ionic interactions. Compared to the Apo-BSA system, the total amino acid–amino acid interactions in the SWCNT-BSA system decreased by 44%. In more detail, a further analysis was conducted on the amino acids with the largest decrease in degree. As illustrated in Figure 5, in the Apo-BSA system, amino acids Lys136, Tyr137, and Tyr139 had the highest degrees, all equal to 8. However, in the SWCNT-BSA system, the degrees of these three amino acids decreased to 2, 5, and 3, respectively, representing reductions of 75%, 37.5%, and 62.5%. This indicates that the degrees of these amino acids were significantly reduced due to the binding of SWCNTs.

MM-GBSA Results Analysis

By calculating the binding free energy, we can more clearly judge the binding ability of single-walled nanotubes with bovine serum albumin (BSA). Then the binding free energy calculation shows that when SWCNT is located in site III, the binding free energy is the lowest, indicating that the system is the most stable, and the interaction between SWCNT and BSA is the strongest. From Table 1, it can be seen that the negative value of free energy indicates that the hydrophobic interaction between SWCNT and BSA promotes the favorable binding. In addition, the contribution of non-polar components is larger, which mainly comes from van der Waals interaction, which further indicates that hydrophobic interaction is the main force for the binding behavior between SWCNT and BSA. The binding affinity further indicates that SWCNT is more stable in site III. SWCNT binds to BSA and has a finely cleaved binding site, which is not in the two classical binding sites of BSA, but in subdomain IB, site III.

Table 1. Binding free energies predicted by MM/GBSA (kcal/mol).

Binding Free Energy Decomposition

To identify the key amino acids contributing significantly to the binding process between SWCNT and BSA, a binding free energy decomposition was performed (Figure 6). The results revealed nine major contributing residues, including Leu115, Lys116, Pro117, Tyr137, Glu140, Ile141, Arg145, Ala157, and Tyr160, all of which are located in the vicinity of the binding pocket. Furthermore, the energy contributions of these nine residues were partitioned into polar and non-polar components (Figure 6). It is noteworthy that residues Trp134, Tyr139, Tyr147, Tyr149, Tyr155, Tyr156, Phe126, Phe133, Phe148, and Phe164, situated within subdomain IB, were also identified as critical residues for the binding interaction. As illustrated in Figure 6, hydrophobic interactions exert a dominant contribution to the binding process, whereas electrostatic interactions contribute minimally, showing almost no impact. Based on these findings, it can be concluded that hydrophobic interaction is the primary driving force governing the BSA-SWCNT binding.

Figure 6. Binding energy analysis of the SWCNT-BSA system. (A) Individual amino acid residue contributions to the binding of SWCNT and BSA. (B) Decomposition of polar (dark gray) and non-polar (light gray) contributions of key residues to the binding free energy. Negative values indicate a favorable contribution to the SWCNT binding process.

Summary

Through molecular dynamics (MD) simulations, binding free energy calculations, and residue interaction network analysis, this study elucidates the interaction mechanism between single-walled carbon nanotubes (SWCNTs) and bovine serum albumin (BSA) from a computational perspective. The primary conclusions are as follows:

1. Binding Mode and Site Identification: MD simulations demonstrate that SWCNTs stably bind within subdomain IB (Site III) of BSA, rather than the classic drug-binding sites. The binding is characterized by a hydrophobic-encapsulated mode.
2. Quantitative Analysis of Interaction Forces: MM/GBSA binding free energy calculations confirm that the binding process is spontaneous. The primary driving force is derived from van der Waals forces (hydrophobic interactions), while the electrostatic contribution is negligible.
3. Elucidation of Structural Perturbation Mechanisms: Residue interaction network analysis reveals that the binding of SWCNTs significantly disrupts the original amino acid interaction network within the binding cavity (with a reduction in total interactions of approximately 44%). This leads to the collapse of the local hydrogen bond network and a decrease in structural stability, providing a molecular basis at the atomic level for the observed changes in the secondary and tertiary structures of BSA.

In summary, this computational study clarifies, with atomic precision, that SWCNTs embed into specific sites of BSA through strong hydrophobic interactions and perturb its structural stability. These findings provide a critical theoretical foundation and molecular mechanism to understand how nanomaterials might interfere with the normal functions of biological macromolecules, thereby inducing potential biological effects.

Related Services

CD ComputaBio provides you with comprehensive services, including but not limited to:

• Protein Molecular Dynamics Simulation
• Binding Free Energy Analysis Service
• Drug Development Service

Connect with Us Anytime!

Studying the interactions between proteins and nanomaterials is essential for understanding biological regulatory mechanisms, and they provide a theoretical basis for the design and discovery of new drug targets. If you require services for inhibitor discovery, please don't hesitate to get in touch with us.

References:

1. Eberhardt J, Santos-Martins D, Tillack AF, Forli S. AutoDock Vina 1.2.0: New docking methods, expanded force field, and python bindings. J Chem Inf Model . 2021; 61(8): 3891–3898.
2. Doncheva NT, Klein K, Domingues FS, Albrecht M. Analyzing and visualizing residue networks of protein structures. Trends Biochem Sci . 2011; 36(4): 179-182.
3. Goddard TD, Huang CC, Ferrin TE. Software extensions to UCSF Chimera for interactive visualization of large molecular assemblies. Structure . 2005; 13(3): 473-482.
4. Hou T, Wang J, Li Y, Wang W. Assessing the performance of the MM/PBSA and MM/GBSA methods. 1. Accuracy of binding free energy calculations based on MD simulations. J Chem Inf Model . 2011; 51(1): 69–82.
5. Genheden S, Ryde U. The MM/PBSA and MM/GBSA methods to estimate ligand-binding affinities. Expert Opin Drug Discov . 2015; 10(5): 449–461.
6. Rastelli G, Di Rienzo A, Degliesposti G, Sgobba M. Fast and accurate predictions of binding free energies using MM-PBSA and MM-GBSA. J Comput Chem . 2010; 31(4): 797–810.
7. Satish L, Millan S, Sasidharan VV, Sahoo H. Molecular level insight into the effect of triethyloctylammonium bromide on the structure, thermal stability, and activity of Bovine serum albumin. Int J Biol Macromol . 2018; 107: 186-193.
8. Brown KF, Crooks MJ. Displacement of tolbutamide, glibencalmide and chlorpropamide from serum albumin by anionic drugs. Biochem Pharmacol . 1976; 25(10): 1175.
9. Kragh-Hansen U. Molecular aspects of ligand binding to serum albumin. Pharmacol Rev . 1981; 33(1): 17-53.
10. Li Z, Hulderman T, Salmen R, et al. Cardiovascular effects of pulmonary exposure to single-wall carbon nanotubes. Environ Health Perspect . 2007; 115(3): 377-382.
11. Legramante JM, Sacco S, Crobeddu P, et al. Changes in Cardiac Autonomic Regulation after Acute Lung Exposure to Carbon Nanotubes: Implications for Occupational Exposure. J Nanomater . 2012; 2012: 1-9.
12. Legramante JM, Valentini F, Magrini A, et al. Cardiac autonomic regulation after lung exposure to carbon nanotubes. Hum Exp Toxicol . 2009; 28(6-7): 369-375.

• Tubulin Polymerization Inhibitor Design
• Protein-ligand Interactions Case Study
• Accelerating Drug Discovery: In Silico Identification and Stability Validation of GSK3B Inhibitors