INTRODUCTION
Introduction about botulism
Botulism, a rare but potentially fatal disease first identified in the 18th century, was notably linked to blood sausage outbreaks in Württemberg, Germany, prompting the introduction of safety measures and its classification as a mandatory reportable disease by 1820 Foods at risk of botulism include sausages, vegetables, olives, cheese, and canned goods, particularly when they become stale due to the presence of the bacteria Clostridium botulinum While the toxin produced by this bacterium often results in mild symptoms, it can also be neurotoxic and life-threatening Botulism can arise from various sources, including bacteria, fungi, algae, and plants, with the neurotoxin Botulinum sometimes found in open wounds or introduced through surgical procedures.
Botulinum neurotoxin (BoNT) proteins
Botulinum neurotoxin, produced by gram-positive anaerobes, is the most lethal poison known to humans, with a fatal dose as low as 1-3 ng/kg Due to its potency, it is classified as a Class A biological warfare agent by the US Centers for Infectious Diseases However, despite its dangers, botulinum neurotoxin is also utilized therapeutically to treat various medical conditions and has received approval from the Food and Drug Administration (FDA) and other regulatory bodies.
Potent inhibition and selective release of Acetylcholine (ACh) can trigger initial symptoms under specific temperature and environmental conditions Botulinum neurotoxins (BoNTs) exhibit remarkable resistance to variations in body temperature, enabling them to remain active for extended periods, potentially lasting months or years Additionally, BoNT can withstand denaturation and proteolysis from stomach acid and digestive enzymes, allowing it to be absorbed through the intestinal wall into the bloodstream while maintaining its original structure The bacteria responsible for producing BoNT include various species.
Clostridium botulinum, along with certain strains of C argentinense, C baratii, and C butyricum, is an anaerobic, rod-shaped bacterium known for producing powerful toxins that lead to botulism symptoms.
Botulinum neurotoxins (BoNT) are classified into seven serotypes (A-G) based on their immunological properties Each serotype has distinct effects on different organisms: BoNT/D is toxic to animals, while BoNT/A, B, E, and F affect humans, and BoNT/C targets avians Despite their differences, these proteins share a similar mechanism of action Once in circulation, BoNT proteins can access various peripheral cell types, binding with high affinity to active nerve synapses They internalize through endocytosis, translocating an enzymatically active protease subunit into the cytoplasm of target cells This combination of actions contributes to the remarkable potency and specificity of BoNTs at active neuromuscular junctions.
[34, 35] In the framework of the thesis, due to the focus of our experimental colleagues, we study BoNT types A and C
Botulinum neurotoxins (BoNTs) are complex toxins consisting of two distinct domains The N-terminal light chain (LC) contains the enzymatic domain, which functions as a zinc metalloprotease, while the C-terminal heavy chain (HC) is made up of two independent functional domains: the translocation domain (HCT) and the receptor-binding domain (HCR) Research indicates that these three domains are structurally independent Previous studies have identified a Clostridium-derived endoprotease as the key factor in cleaving BoNT/A to form the di-chain toxin.
46] where cleavage removed ten amino acids at the junction between the LC and the
HC yielding an LC of ~438 amino acids, which is considered the active form of the
The light chain (LC) of botulinum neurotoxin type A plays a crucial role in its toxicity, with research indicating that the reduction of toxicity in various BoNT derivatives produced by clostridia in an unnicked form highlights the significance of the nicking process Consequently, the light chain is commonly targeted by biosensors for the detection of this potent neurotoxin Figure 1.1 illustrates the structure of the light chain of botulinum neurotoxin type A.
Figure 1.1: The three dimensions structure for the light chain BoNT type A (download from the Protein Data bank with code ID 3DDA)
Aptamer
Nanotechnology and aptamer technology are poised to revolutionize the rapid detection of biological elements, particularly botulinum neurotoxin (BoNT) Aptamers, which are short, single-stranded DNA or RNA molecules, can selectively bind to various targets, including proteins, peptides, carbohydrates, small molecules, toxins, and live cells Their unique ability to adopt diverse shapes, such as helices and single-stranded loops, enhances their versatility, allowing for high selectivity and specificity in binding Additionally, aptamers can be easily chemically modified, further expanding their potential applications.
In 2015, J Shi introduced a method for detecting BoNT strain A using graphene materials and the fluorescence quenching effect (FRET) This approach enables autonomy in aptamer design This thesis aims to explore the interaction between aptamers and proteins, proposing aptamers as highly specific targets for recognizing botulinum neurotoxin.
Dual detection biosensor of BoNT type A and type C
In Vietnam, various experimental groups are advancing multidisciplinary research on the application of (nano-)materials in biomedicine, particularly in biosensor manufacturing Notable teams include those led by Assoc Dr Nguyen Hoang Nam and Dr Luu Manh Quynh from VNU University of Science, Assoc Dr Mai Anh Tuan from the Nanotechnology Laboratory at the Ministry of Science and Technology, and Dr Nguyen Van Chuc from the Department of Nano Carbon Materials at the Institute of Materials Science, Vietnam Academy of Sciences Their research emphasizes the use of materials like nano carbon structures and graphene in biomedical and environmental applications, with aptamers playing a crucial role in molecular recognition.
A biosensor typically consists of three key components: first, the recognition element, which operates on a "key & lock" principle to specifically identify the target substance, such as botulinum neurotoxin (BoNT); second, the signal transduction mechanism that transforms the physiological signal generated during recognition into a readable output; and third, the signal processing unit, which assesses the output signal and correlates it with the physical parameters of the subject being analyzed.
Experimental studies must thoroughly detail the fabrication steps of sensor elements to accurately assess their quality In sensors that utilize aptamers to detect the presence of botulinum neurotoxin (BoNT) in samples, the binding interaction between the aptamer and BoNT is crucial To develop effective aptamers, researchers commonly employ the SELEX (Systematic Evolution of Ligands by Exponential Enrichment) method.
The selection of the most suitable aptamer strain for recognizing BoNT is complex due to the multitude of potential aptamer chains This study employs a theoretical approach, utilizing Molecular Dynamics (MD) simulations to elucidate the interaction mechanisms between aptamers and BoNT, thereby aiding in the identification of promising aptamers for experimental validation Our future goal is to design and develop biosensors capable of dual detection for BoNT types A and C, with theoretical findings being shared and collaborated on with Dr Luu Manh Quynh's experimental group at the Center for Materials Science, VNU University of Science.
METHODOLOGY
Molecular dynamics (MD) simulation
In essence, molecular dynamics are the numerical solution of the Newton’s equations of motion is solved for a system of N atoms
𝜕𝑡 2 = 𝐹 𝑖 , 𝑖 = 1 … 𝑁 (1) where the force on each particle is minus the derivative of the total potential energy:
In computational simulations, time is discretized into small intervals (typically 1-2 femtoseconds for biomolecules), allowing for simultaneous resolution of equations at each time step The system's trajectory is represented by the coordinates as a function of time, beginning from an initial configuration and progressing towards equilibrium over a defined period Macroscopic properties are derived by averaging the values from the equilibrium trajectory recorded in the output file.
In this study, we utilize the GROMACS software package to conduct molecular dynamics simulations of the aptamer-BoNT protein complex The initial configuration is derived from the molecular docking process, selecting the optimal docking pose with the lowest free energy.
All simulations were conducted using GROMACS 2020.4 with the Amber99sb-ILDN force field Each system's energy was minimized to 1000.0 kJ/mol/nm through the steepest descent algorithm Following this, the constrained protein and ligand systems underwent a 10 ns equilibration at constant pressure and temperature.
Molecular dynamics (MD) simulations were conducted for 100 nanoseconds at 300 K and 1 atm using a semi-isotropic pressure approach The NPT ensemble was maintained with a Nose Hoover thermostat and Parrinello Rahman pressure coupling A leap-frog algorithm was employed as the integrator, with a time step of 2 femtoseconds, which is standard for biological molecules.
The structures of botulinum neurotoxin type A (BoNT/A) and type C (BoNT/C) were obtained from the Protein Data Bank, identified by IDs 3DDA and 2QN0, with resolutions of 1.5Å and 1.75Å, respectively Both protein light chains consist of 430 residues, as detailed in Table 2.1.
Table 2.1: The primary structure of BoNT type A and Type C light chain Both are 430 residues long The residues of coordinates with the zinc ion are denoted by bold letter
MPITI NNFNY SDPVD NKNIL YLDTH LNTLA NEPEK AFRIT GNIWV IPDRF
MPFVN KQFNY KDPVN GVDIA YIKIP NAGQM QPVKA FKIHN KIWVI PERDT
SRNSN PNLNK PPRVT SPKSG YYDPN YLSTD SDKDT FLKEI IKLFK RINSR
FTNPE EGDLN PPPEA KQVPV SYYDS TYLST DNEKD NYLKG VTKLF ERIYS
EIGEE LIYRL STDIP FPGNN NTPIN TFDFD VDFNS VDVKT RQGNN WVKTG
TDLGR MLLTS IVRGI PFWGG STIDT ELKVI DTNCI NVIQP DGSYR SEELN
SINPS VIITG PRENI IDPET STFKL TNNTF AAQEG FGALS IISIS PRFML
LVIIG PSADI IQFEC KSFGH EVLNL TRNGY GSTQY IRFSP DFTFG FEESL
TYSNA TNDVG EGRFS KSEFC MDPIL ILMHE LNHAM HNLYG IAIPN DQTIS
EVDTN PLLGA GKFAT DPAVT LAHEL IHAGH RLYGI AINPN RVFKV NTNAY
SVTSN IFYSQ YNVKL EYAEI YAFGG PTIDL IPKSA RKYFE EKALD YYRSI
YEMSG LEVSF EELRT FGGHD AKFID SLQEN EFRLY YYNKF KDIAS TLNKA
AKRLN SITTA NPSSF NKYIG EYKQK LIRKY RFVVE SSGEV TVNRN KFVEL
KSIVG TTASL QYMKN VFKEK YLLSE DTSGK FSVDK LKFDK LYKML TEIYT
YNELT QIFTE FNYAK IYNVQ NRKIY LSNVY TPVTA NILDD NVYDI QNGFN
EDNFV KFFKV LNRKT YLNFD KAVFK INIVP KVNYT IYDGF NLRNT NLAAN
IPKSN LNVLF MGQNL SRNPA LRKVN PEPLV
FNGQN TEINN MNFTK LKNFT GLFEH HHHHH
The zinc finger structure of the botulinum neurotoxin (BoNT) light chain is crucial for its catalytic function, with type A (left) and type C (right) displaying distinct characteristics The zinc ion plays a vital role by coordinating with the target substrate protein, enhancing the enzyme's activity Notably, in the BoNT type A light chain, the fourth residue is derived from the human SNAP-25 protein segment, highlighting its significance in the neurotoxin's mechanism.
Notice that, both BoNT type A and type C proteins contain one zinc-finger structure
The structure of the BoNT type C light chain features a zinc ion that coordinates with four residues: His229, His233, Glu230, and Glu269 In contrast, the BoNT type A light chain has a zinc ion coordinating with three residues—His227, His223, and Glu262—along with residue Gln197 from the human SNAP-25 segment This coordination is crucial for maintaining the stability of the zinc-finger structure within the aptamer-BoNT complex Additionally, both BoNT and clostridial neurotoxins (CNT) are classified as zinc-dependent endopeptidases, making the stabilization of the zinc finger vital for the catalytic properties of CNT.
To ensure the stability of the zinc finger structure, it is essential to accurately assign the protonation states of the amino acids coordinating the zinc ion, as highlighted in our previous Biochemistry publication [61] While GROMACS software can automate this process, a thorough manual inspection is crucial to confirm that the zinc-finger structure is correctly modeled.
Experimentally, depending on the type of BoNT, there are 10-20 candidate aptamers for their binding In this work, we pick the best 5 candidate aptamers for each type
This approach simplifies the problem given our limited computational resources and emphasizes understanding the primary binding mechanism of the aptamer-BoNT complex Specifically, the nucleic acid sequences, or primary structures, of the aptamers studied in this research are detailed in Table 2.2.
This study investigates a selection of five aptamers for each type of botulinum neurotoxin (BoNT), chosen from a curated list of experimental aptamers that demonstrate strong binding affinity to their respective target proteins.
Residue sequence Number of residues
49 Type A with proteins Therefore, experiments have shown that the binding affinity are comparable to each other
2.1.4 Method for building secondary structures of Aptamers
An essential step in constructing the three-dimensional structure of aptamers is identifying the pairing of nucleic acid residues Unlike double-stranded DNA, where complementary strands form a stable double helix, single-stranded DNA must fold upon itself, leading to imperfect pairings This results in short double helical segments interspersed with structures like hairpins and loops Predicting the secondary structure of single-stranded DNA is an active area of research, with various software tools available In this study, we will utilize MFold, a free and well-established webserver for nucleic acid secondary structure prediction Although MFold does not account for complex knot structures, this limitation is not significant for our short aptamer sequences, as knots are rarely thermodynamically stable due to high configurational entropy penalties.
The three-dimensional structure of aptamers is constructed using the outputs from MFold and the RNAcomposer webserver This process involves building the orientation of side groups, stems, and loops based on a database of experimentally available templates To ensure accuracy, the structure undergoes energetic optimization to eliminate overlaps of side groups and address torsional angle space and steric constraints Typically, the optimization process for most aptamers requires around 1000 cycles, with example outputs provided in the Appendix.
Docking of Molecular biomolecules
Molecular docking is the prediction how small molecules, such as drugs and substrates, bind to a 3D structure biomolecular
Most docking software operates automatically, efficiently predicting bound conformations and calculating binding energies using a semiempirical free energy force field They utilize a grid-based method to explore the extensive conformational space of a ligand around a protein, allowing for rapid evaluation of binding energies for various trial conformations This approach resembles a Monte-Carlo simulated annealing method, leading to slightly different optimal poses in different docking runs Typically, 5-10 molecular docking procedures are conducted for each aptamer-protein complex, with the one exhibiting the lowest binding free energy being selected.
In this thesis, we employ the MOE software to dock aptamers to BoNT proteins MOE utilizes a docking methodology akin to the well-known AutoDock program, but it is specifically optimized for larger ligands like aptamers.
Interaction free energy estimation by MM-PBSA method
Molecular mechanics Poisson−Boltzmann surface area (MM-PBSA) approach has been widely used to fast compute interaction energies from molecular dynamics outputs, especially, for biomolecular complexes
Generally, the binding free energy is calculated as
∆𝐺 𝑏𝑖𝑛𝑑 = 𝐺 𝑐𝑜𝑚𝑝𝑙𝑒𝑥 − (𝐺 𝑝𝑟𝑜𝑡𝑒𝑖𝑛 + 𝐺 𝑙𝑖𝑔𝑎𝑛𝑑 ) (3) where Gcomplex is the total free energy of the protein−ligand complex and Gprotein and
Gligand represents the total free energies of the isolated protein and ligand in a solvent To enhance efficiency, it is beneficial to segment the calculations based on the thermodynamic cycle illustrated in Figure 2.3.
Figure 2.3: Thermal dynamics cycle for calculation of ligand-protein binding energy Therefore, the binding free energy ΔGbind,solv is:
Solvation free energies are determined by solving the linearized Poisson-Boltzmann equation for three states, which provides the electrostatic contribution to solvation free energy, combined with an empirical term for hydrophobic effects The equation for solvation free energy is given by ∆𝐺 𝑠𝑜𝑙𝑣 0 = 𝐺𝑒𝑙𝑒𝑐𝑡𝑟𝑜𝑠𝑡𝑎𝑡𝑖𝑐,𝑐 0 − 𝐺𝑒𝑙𝑒𝑐𝑡𝑟𝑜𝑠𝑡𝑎𝑡𝑖𝑐,𝑐=1 0 + ∆𝐺 𝑛𝑜𝑛−𝑝𝑜𝑙𝑎𝑟 0 The vacuum free energy, ΔGVacuum, is calculated by averaging the interaction energy between the receptor and ligand, adjusting for entropy changes during binding if necessary, expressed as ∆𝐺 𝑣𝑎𝑐𝑢𝑢𝑚 0 = ∆𝐸𝑚𝑜𝑙𝑒𝑐𝑢𝑙𝑎𝑟 𝑚𝑒𝑐ℎ𝑎𝑛𝑖𝑐𝑠 0 − 𝑇 ∆𝑆𝑛𝑜𝑚𝑎𝑙 𝑚𝑜𝑑𝑒 𝑎𝑛𝑎𝑙𝑦𝑠𝑖𝑠 0.
Normal mode analysis can provide entropy contributions for three species; however, these contributions can often be disregarded when comparing states with similar entropy, such as two ligands binding to the same protein This is primarily due to the high computational cost of normal mode analysis compared to MM-PBSA, along with the significant standard error that can lead to uncertainty in the results.
The MM/PBSA method accounts for various interactions among ligands, proteins, solvents, and ions, with electrostatic interactions being the only long-range type explicitly considered In contrast, short-range interactions such as van der Waals forces, excluded volume effects, and hydrophobicity are incorporated into the semi-empirical expression for the non-polar term in the MM/PBSA calculations.
The equation ∆𝐺 𝑛𝑜𝑛−𝑝𝑜𝑙𝑎𝑟 0 = 𝛾 × 𝑆𝐴𝑆𝐴 illustrates the relationship between the non-polar free energy change and the Solvent Accessible Surface Area (SASA) The surface tension factor γ is established at 0.0227 kcal/mol/A², based on experimental measurements Notably, the uncertainty associated with this factor is significantly smaller compared to the larger electrostatic term.
In this thesis, the average interaction energies between aptamers and proteins are calculated using an ensemble of uncorrelated snapshots gathered from molecular dynamics (MD) trajectories, with snapshots taken every 10 nanoseconds for MM-PBSA analysis This interval is generally greater than the NPT relaxation time.
Flowchart of the computational pipeline
Create simulation box, add waters and ions
Analyses and pick optimal docking pose
NVT and NPT ensemble equilibrations
Molecular dynamics simulations for 100ns
Analyses: RMSD, RMSF, H-bond, MM/PBSA
3D structure of BoNT from Protein Data Bank
RESULTS AND DISCUSSION
Docking of aptamers with BoNT type A and type C proteins
Once the proteins and aptamers are meticulously designed with all key substructures identified, the next step involves docking the aptamers to the targeted proteins The binding free energy results for the lowest energy configurations are presented in Table 3.1, while the three-dimensional structures of each individual aptamer-BoNT complex are illustrated in Figure 3.2.
Table 3.1: Binding free energy of the docking pose with the lowest S score, configuration and placement energy
Aptamer Complex with BoNT Type C Complex with BoNT Type A C5
Figure 3.2: The three-dimensional structures of the aptamer – BoNT complex with the lowest S score
From these docking results, upon closer look at the energy values as well as structural aspects of the complexes, we have the following observations:
All aptamers exhibit strong binding affinity to both types of botulinum neurotoxin (BoNT), with binding free energies ranging from -13 to -30 kcal/mol, confirming their effectiveness as candidates for targeting these proteins Notably, these aptamers demonstrate cross-reactivity, with those selected for BoNT/A also binding well to BoNT/C and vice versa This cross-binding is not surprising due to the similarities in enzymatic functions and structural characteristics of the two protein types, which share common features such as substrate-binding zinc-finger structures and other domain structures despite differences in their protein sequences.
In optimal docking complexes like aptamer C4 – BoNT/C, the aptamer minimally interacts with the protein at a few terminal residues, yet exhibits a strong docking free energy of -24 kcal/mol This highlights a limitation of static ligand-receptor docking, which fails to account for the configurational changes of the molecules upon complexation While this limitation is less critical for small drug molecules due to their rigidity, it is significant for larger ligands like aptamers, where flexibility must be considered for accurate results Despite this, the docking pose can provide a valuable starting point for further dynamical optimization through molecular dynamics simulations.
We selected 10 aptamers based on experimental suggestions, and five of them—A2, A3, A4, C4, and C5—demonstrated strong binding energy to both BoNT/C and BoNT/A These aptamers will be utilized for molecular dynamic simulations of the aptamer-BoNT complexes, each simulated for 100 ns to facilitate structural optimization and analyze individual interactions between the aptamers and proteins, all while remaining within the computational limits of our HPC In collaboration with Dr Luu Manh Quynh's experimental group at the Center for Materials Science, VNU University of Science, we aim to conduct longer simulations on a specific aptamer to explore clustering information, phase transitions, and the dynamic properties of the complexes.
Molecular dynamics simulation results
In this study, we focus on the structural and energetic properties of five aptamers—A2, A3, A4, C4, and C5—selected for their strong binding free energy with both types of BoNT proteins Each complex undergoes a 100ns simulation under physiological conditions, specifically at 1 atm, 300K, and 150 mM NaCl, with additional Na or Cl ions to neutralize the system The simulations utilize an explicit solvent model, employing the TIP3P model for water molecules.
3.3.1 Root mean square deviation and equilibrium analysis
To investigate the equilibrium approach of each complex from its optimal docking pose, we calculate the root-mean-square deviation (RMSD) for each component molecule both in relation to itself and to other molecules Figure 3.3 presents the RMSD values, detailing the protein's deviation from itself and the aptamer's deviation in relation to the protein.
Figure 3.3: The RMSD values of protein (column 2) and aptamer (column 3), both with respect to the protein
From these analyses, we have the following preliminary conclusion:
- In all the systems simulated, the proteins are very stable, with RMSD values quick saturate at about 1.5Å after a few nanoseconds This is much smaller than typical
The RMSD of 5Å for the isolated protein in a water solution indicates that complexation with aptamers significantly stabilizes the protein, limiting its deviation from the experimentally determined low-temperature configuration.
- Aptamers A3 and A4 shows very similar deviation for both types of BoNT proteins This is in agreement with free energy of docking
Aptamer A2 demonstrates a high specificity for BoNT/A proteins, exhibiting a minimal deviation of just 1nm In contrast, when interacting with BoNT/C, the root mean square deviation (RMSD) significantly increases to 4.5nm over a duration of 100ns, which aligns with the free energy results observed during docking studies.
Aptamer C4 exhibits a stronger binding affinity for BoNT type A, while C5 shows a preference for BoNT type C, indicating contrasting trends in RMSD However, these differences fall within the margin of uncertainty, suggesting that the docking results may not fully capture the binding dynamics.
The protein complexes measure approximately 8-10nm, exhibiting a small RMSD value of 0.10-0.15nm, which signifies their stability In contrast, the RMSD values for aptamers relative to the protein range from 1-3nm, highlighting a significant structural relaxation from their initial docked state Given the flexibility and size of aptamers, traditional static docking methods fall short in accurately representing the binding dynamics between aptamers and proteins.
3.3.2 Root mean square fluctuation and stability analysis
Following equilibrium analyses, the stability analysis is conducted by calculating the root-mean-square fluctuations (RMSF) of each residue around the average structure In this process, the initial 20 ns of simulation data is excluded, and the remaining 80 ns is utilized for averaging Residues that interact with BoNT proteins typically exhibit lower RMSF values, whereas residues located at free ends and loops tend to display higher RMSF values.
Aptamer RMSF for each of the aptamer residues
Figure 3.4: Root mean square fluctuation of aptamer in complexation with BoNT/A
(violet color) and BoNT/C (cyan color)
The Root Mean Square Fluctuation (RMSF) of the proteins is intentionally omitted for clarity, as the Root Mean Square Deviation (RMSD) analyses have already demonstrated the proteins' stability within the complex Including RMSF analyses would yield minimal additional insights This thesis aims to explore the potential of various aptamers as candidates for binding to both types of Botulinum Neurotoxin (BoNT).
From Figure 3.4, the following observations are made regarding the RMSF values of the aptamers:
Aptamer A2 demonstrates a higher specificity in binding to BoNT/A compared to BoNT/C, as indicated by a significantly lower root mean square fluctuation (RMSF) during the interaction with BoNT/A This finding is consistent with the results from root mean square deviation (RMSD) and docking analyses.
- Aptamer A3 and A4 show a similar stability upon binding to BoNT/A and BoNT/C This is in agreement with RMSD and docking analyses
Aptamers C4 and C5 exhibit comparable stability when binding to BoNT/A and BoNT/C, with C5 showing slightly greater stability with BoNT/C, consistent with docking results and RSMD analysis Accurate assessment of hydrogen bonds would necessitate quantum mechanical calculations, which are impractical for large complexes like these Instead, we employed standard geometric criteria for hydrogen bond classification, which may lead to an overestimation of hydrogen bonds due to the potential for a receptor atom to engage in multiple bonds Nonetheless, the qualitative conclusions remain valid regardless of the hydrogen bond analysis method used.
The results of calculation of the number of hydrogen bonds between aptamer and BoNT protein for each system are listed in the Figure 3.5
Aptamer The number of Hydrogen bonds
The analysis of hydrogen bonds between the aptamer and BoNT proteins over time reveals that there are more bonds formed with BoNT/A compared to BoNT/C This finding is consistent with the results from docking and RMSF analyses, as well as the RMSD analyses, which fall within the margin of uncertainty.
Among the aptamers, aptamer A4 has the highest number of hydrogen bonds This makes them the most suitable candidate for further experimental study
3.3.4 Binding free energy from MM/PBSA calculation
In our final analysis, we utilized the MM/PBSA method to calculate the binding free energy of the aptamer with the protein The dielectric constant for water was set to 80, while that for the protein environment was set to 2 We employed the mean-field Poisson Boltzmann method to calculate the polar and electrostatic energy, with the results detailed in Table 3.2.
Table 3.2 presents the binding free energy (measured in kJ/mol) of the aptamer when interacting with BoNT/A or BoNT/C, calculated using the MM/PBSA method The table includes columns for van der Waals energy, electrostatic energy, solvation energy, solvent-accessible surface area (SASA) energy, and the total binding energy.
System Van der Waal energy (kJ/mol)
Polar solvation energy (kJ/mol)
The binding free energy calculated from molecular dynamics (MD) simulations is consistently negative across all systems, indicating effective protein binding However, unlike docking results, there are significant variations in the MD simulation energies This discrepancy arises from the relaxation of the complex from its static docking pose Unlike small drug molecules, aptamers are larger ligands with greater degrees of freedom, and docking fails to account for the flexible aptamer structure's adaptation to the protein upon complex formation.
The key finding from Table 3.2 is that aptamer A4 exhibits the strongest binding energy to both types of Botulinum Neurotoxin (BoNT) Its binding to BoNT/A is the most robust, as expected for the target protein, while its binding to BoNT/C is comparable and may even exceed that of aptamers C4 and C5 This computational analysis indicates that aptamer A4 has significant potential for use in biosensors designed to detect both BoNT/A and BoNT/C.
Discussion
The integration of docking calculations, RMSD and RMSF structural analyses, hydrogen bond assessments, and MM/PBSA energetic evaluations reveals the key interactions and specificity of aptamers with proteins The qualitative characteristics of these complexes are effectively represented in the docking results, highlighting molecular docking as a valuable pre-screening tool among a vast array of aptamers, despite its notable limitations.
Both proteins exhibit remarkable stability, indicated by low RMSD values relative to themselves and minimal RMSF of the amino acid residues The RMSD values are significantly lower than those typically observed for isolated proteins in aqueous solutions, highlighting the enhanced stability and compactness achieved through aptamer complexation This observation further supports the notion that the optimal docking pose is an excellent starting point for molecular dynamics simulations.
Aptamer A4 is the leading candidate for biosensor applications targeting both types of Botulinum neurotoxin Initially designed for selectivity towards BoNT/A, it demonstrates comparable binding properties to BoNT/C across various structural and energetic analyses.
Aptamer A2 exhibits comparable binding affinity to both BoNT/A and BoNT/C according to docking results; however, molecular dynamics simulations reveal significant differences in binding energies and stabilities for these proteins Notably, the energy values for A2 are considerably lower than those of other aptamers, with the A2-BoNT/C complex failing to reach equilibrium during the simulation timeframe This discrepancy indicates that the docking results for this aptamer may be unreliable, and it is through the insights gained from MD simulations that we can eliminate this aptamer from further experimental studies.
Aptamer A3 exhibits a binding affinity to both types of BoNT comparable to that of aptamer A4; however, it demonstrates significantly weaker hydrogen bonds and binding energy Consequently, we do not recommend aptamer A3 for further experimental investigation.
Aptamers C4 and C5 exhibit a stronger preference for their target, BoNT/C, compared to the cross-target BoNT/A, although this distinction is less pronounced than the preference shown by aptamer A2 for BoNT/A Notably, aptamer C5 stands out as a promising candidate for experimental investigations aimed at detecting BoNT/C For cross-type detection, we recommend aptamer A4, which is commercially available and provides a cost-effective option for biosensor development.
The MM/PBSA calculations for the C5 aptamer yield unexpected results in docking studies, indicating a preference for binding to BoNT/A, despite being designed for BoNT/C However, Table 3.2 confirms that the aptamer actually favors BoNT/C, emphasizing the limitations of molecular docking methods when applied to large, flexible ligands like aptamers.
Aptamer A4 exhibits strong binding to BoNT/C, likely due to its similar secondary structure to aptamer C5, as illustrated in Figure 3.1 Both aptamers feature two hairpins and four short double helix segments Future research will further explore this structural similarity.
This thesis presents preliminary computational work aimed at enhancing the experimental design of biosensors for detecting Botulinum Neurotoxin types A and C, as researched by Dr LUU Manh Quynh at the Center for Materials Sciences, VNU University of Science We have identified five candidate aptamers for each type of BoNT proteins Through molecular docking and molecular dynamics simulations of their interactions with the target proteins, as well as cross-binding assessments, we aim to guide the selection and synthesis of aptamers, ultimately reducing time, labor, and material costs in the experimental process.
Aptamers are large protein ligands with considerable flexibility, making standard docking methods inadequate for accurately capturing the structural changes that occur in the protein-aptamer complex during binding To thoroughly investigate binding affinity, molecular dynamics simulations are essential However, our findings indicate that docking poses can effectively serve as initial configurations for further analysis.
Our study reveals that during molecular dynamics (MD) simulation, the protein (BoNT/A or BoNT/C) shows minimal deviation from its original configuration, measuring only 1-2Å Additionally, the presence of aptamers does not interfere with the protein's structure; instead, they contribute to its enhanced stability.
In turn, this allows aptamers to be an excellent candidate for biosensor for detection
All aptamers demonstrate good binding affinity to their respective targets, with a notable preference for their designed target proteins; for instance, aptamer A4 shows strong binding to both BoNT/A and BoNT/C, characterized by significant hydrogen bond formation and favorable binding free energy This suggests that aptamer A4 is a promising candidate for use in biosensors for cross detection However, for specific detection of BoNT/C, the C5 aptamer emerges as a superior choice over A4 Interestingly, the C5 aptamer shares a similar secondary structure with A4, featuring two hairpins and four short helical segments, which may account for A4's effective binding to BoNT/C as well.
In the future, we aim to conduct a more detailed investigation of aptamers A4 and C5, which share similarities in their secondary structures and effectively bind to BoNT/C, with A4 also showing a strong affinity for its preferred target, BoNT/A Our research will focus on elucidating the binding sites between these aptamers and the proteins, as well as assessing their stability This exploration may unveil critical insights into the molecular mechanisms of binding, aiding in the optimization and design of new aptamers targeting Botulinum neurotoxin proteins and a wider range of clostridial neurotoxins.
[1] Peel R N (1989) Botulism The Organism, Its Toxins, The Disease Journal of
[2] Low PA (2002) Autonomic neuropathies Curr Opin Neurol 15:605–609
[3] Poulain B, Popoff MR, Molgo J (2008) How do the botulinum neurotoxins block neurotransmitter release: from botulism to the molecular mechanism of action
Botulinum J 1:14–87 http://dx.doi.org/10.1504/TBJ.2008.018951
[4] Tacket CO, Rogawski MA (1989) Botulism In: Simpson LL (ed) Botulinum neurotoxin and Tetanus toxin Academic Press, San Diego, 351–378
[5] Gill DM (1982) Bacterial toxins: a table of lethal amounts Microbiol Rev 46:86–
[6] Bhidayasiri R, Truong DD (2005) Expanding use of botulinum toxin J Neuro Sci
[7] Bhidayasiri R, Truong DD (2008) Evidence for effectiveness of botulinum toxin for hyperhidrosis J Neural Transm 115:641–645 https://doi.org/10.1007/s00702-007- 0812-7
[8] Caya, J G., Agni, R., & Miller, J E (2004) Clostridium botulinum and the clinical laboratorian: a detailed review of botulism, including biological warfare ramifications of botulinum toxin Archives of pathology & laboratory medicine, 128(6), 653–662 https://doi.org/10.5858/2004-128-653-CBATCL
I don't know!
I don't know!
[11] Foster KA (2004) The analgesic potential of clostridial neurotoxin derivatives
Expert Opin Investig Drugs 13:1437–1443 https://doi.org/10.1517/13543784.13.11.1437
[12] Foster KA (2009) Engineered toxins: new therapeutics Toxicon 54:587–592 https://doi.org/10.1016/j.toxicon.2009.01.037
[13] Foster, K A., Bigalke, H., & Aoki, K R (2006) Botulinum neurotoxin - from laboratory to bedside Neurotoxicity research, 9(2-3), 133–140 https://doi.org/10.1007/BF03033931
[14] AMBACHE N (1951) A further survey of the action of Clostridium botulinum toxin upon different types of autonomic nerve fibre The Journal of physiology, 113(1), 1–17 https://doi.org/10.1113/jphysiol.1951.sp004551
[15] BROOKS V B (1956) An intracellular study of the action of repetitive nerve volleys and of botulinum toxin on miniature end-plate potentials The Journal of physiology, 134(2), 264–277 https://doi.org/10.1113/jphysiol.1956.sp005642
[16] BURGEN, A S., DICKENS, F., & ZATMAN, L J (1949) The action of botulinum toxin on the neuro-muscular junction The Journal of physiology, 109(1-2), 10–24 https://doi.org/10.1113/jphysiol.1949.sp004364
[17] Dressler, D., Eleopra, R (2006) Clinical use of non-a botulinum toxins: botulinum toxin type B neurotox res 9 , 121–125 https://doi.org/10.1007/BF03033929
[18] Edmunds CW, KeiperGF Jr (1924) Further studies on the action of botulinus toxin JAMA 83:495–501
[19] Kao, I., Drachman, D B., & Price, D L (1976) Botulinum toxin: mechanism of presynaptic blockade Science (New York, N.Y.), 193(4259), 1256–1258 https://doi.org/10.1126/science.785600
[20] Merz, B., Bigalke, H., Stoll, G., & Naumann, M (2003) Botulism type B presenting as pure autonomic dysfunction Clinical autonomic research : official journal of the Clinical Autonomic Research Society, 13(5), 337–338 https://doi.org/10.1007/s10286-003-0118-2
[21] Aoki KR (2002) Physiology and pharmacology of therapeutic botulinum neurotoxins Curr Probl Dermatol 30:107–116 https://doi.org/10.1159/000060683
[22] Brashear A (2010) Botulinum toxin type A: Exploring new indications Drugs of today (Barcelona, Spain : 1998), 46(9), 671–682 https://doi.org/10.1358/dot.2010.46.9.1524356
[23] Brin M F (2009) Basic and clinical aspects of BOTOX Toxicon : official journal of the International Society on Toxinology, 54(5), 676–682 https://doi.org/10.1016/j.toxicon.2009.03.021
[24] Carruthers, J., & Carruthers, A (2009) Botulinum toxin in facial rejuvenation: an update Dermatologic clinics, 27(4), 417–v https://doi.org/10.1016/j.det.2009.08.001
[25] Dressler D (2010) Comparing Botox and Xeomin for axillary hyperhidrosis J Neural Transm 117:317–319.https://doi.org/10.1007/s00702-010-0372-0
[26] Dutton, J J., & Fowler, A M (2007) Botulinum toxin in ophthalmology Survey of ophthalmology, 52(1), 13–31 https://doi.org/10.1016/j.survophthal.2006.10.003
[27] Silberstein, S D., Blumenfeld, A M., Cady, R K., Turner, I M., Lipton, R B., Diener, H C., Aurora, S K., Sirimanne, M., DeGryse, R E., Turkel, C C., & Dodick,
D W (2013) OnabotulinumtoxinA for treatment of chronic migraine: PREEMPT 24- similar asaccharolytic clostridia Journal of clinical microbiology, 26(11), 2447–2449 https://doi.org/10.1128/jcm.26.11.2447-2449.1988
[30] Peck M W (2009) Biology and genomic analysis of Clostridium botulinum Advances in microbial physiology, 55, 183–320 https://doi.org/10.1016/S0065-2911(09)05503-9
[31] Call JE, Cooke PH, Miller AJ (1995) In situ characterization of Clostridium botulinum neurotoxin synthesis and export J Appl Bacteriol 79:257–263 https://doi.org/10.1111/j.1365-2672.1995.tb03135.x
[32] Karalewitz, A P., Kroken, A R., Fu, Z., Baldwin, M R., Kim, J J., & Barbieri, J
T (2010) Identification of a unique ganglioside binding loop within botulinum neurotoxins C and D-SA Biochemistry, 49(37), 8117–8126 https://doi.org/10.1021/bi100865f
[33] Molecular aspects of botulinum neurotoxin (2014) New York: Springer
[34] Brunger, A T., & Rummel, A (2009) Receptor and substrate interactions of clostridial neurotoxins Toxicon : official journal of the International Society on
Toxinology, 54(5), 550–560 https://doi.org/10.1016/j.toxicon.2008.12.027
[35] Lalli, G., Bohnert, S., Deinhardt, K., Verastegui, C., & Schiavo, G (2003) The journey of tetanus and botulinum neurotoxins in neurons Trends in microbiology, 11(9), 431–437 https://doi.org/10.1016/s0966-842x(03)00210-5