In Silico Identification of a Potential TNF-Alpha Binder Using a Structural Similarity: A Potential Drug Repurposing Approach to the Management of Alzheimer's Disease

Introduction Alzheimer's disease (AD) is a neurodegenerative disorder with no conclusive remedy. Yohimbine, found in Rauwolfia vomitoria, may reduce brain inflammation by targeting tumour necrosis factor-alpha (TNFα), implicated in AD pathogenesis. Metoserpate, a synthetic compound, may inhibit TNFα. The study is aimed at assessing the potential utility of repurposing metoserpate for TNFα inhibition to reduce neuronal damage and inflammation in AD. The development of safe and effective treatments for AD is crucial to address the growing burden of the disease, which is projected to double over the next two decades. Methods Our study repurposed an FDA-approved drug as TNFα inhibitor for AD management using structural similarity studies, molecular docking, and molecular dynamics simulations. Yohimbine was used as a reference compound. Molecular docking used SeeSAR, and molecular dynamics simulation used GROMACS. Results Metoserpate was selected from 10 compounds similar to yohimbine based on pharmacokinetic properties and FDA approval status. Molecular docking and simulation studies showed a stable interaction between metoserpate and TNFα over 100 ns (100000 ps). This suggests a reliable and robust interaction between the protein and ligand, supporting the potential utility of repurposing metoserpate for TNFα inhibition in AD treatment. Conclusion Our study has identified metoserpate, a previously FDA-approved antihypertensive agent, as a promising candidate for inhibiting TNFα in the management of AD.

Introduction.Alzheimer's disease (AD) is a neurodegenerative disorder with no conclusive remedy.Yohimbine, found in Rauwolfia vomitoria, may reduce brain inflammation by targeting tumour necrosis factor-alpha (TNFα), implicated in AD pathogenesis.Metoserpate, a synthetic compound, may inhibit TNFα.The study is aimed at assessing the potential utility of repurposing metoserpate for TNFα inhibition to reduce neuronal damage and inflammation in AD.The development of safe and effective treatments for AD is crucial to address the growing burden of the disease, which is projected to double over the next two decades.Methods.Our study repurposed an FDA-approved drug as TNFα inhibitor for AD management using structural similarity studies, molecular docking, and molecular dynamics simulations.Yohimbine was used as a reference compound.Molecular docking used SeeSAR, and molecular dynamics simulation used GROMACS.Results.Metoserpate was selected from 10 compounds similar to yohimbine based on pharmacokinetic properties and FDA approval status.Molecular docking and simulation studies showed a stable interaction between metoserpate and TNFα over 100 ns (100000 ps).This suggests a reliable and robust interaction between the protein and ligand, supporting the potential utility of repurposing metoserpate for TNFα inhibition in AD treatment.Conclusion.Our study has identified metoserpate, a previously FDAapproved antihypertensive agent, as a promising candidate for inhibiting TNFα in the management of AD.

Introduction
Alzheimer's disease (AD) is a progressive neurodegenerative disorder that is prevalent among the elderly population and is the leading cause of dementia [1,2].The prevalence of AD is increasing steadily and is predicted to double in the next 20 years [3].Its pathogenesis encompasses the accumulation of beta-amyloid plaques, neurofibrillary tangles, and neuronal loss in the brain.It results in memory and cognitive function deterioration, which affects the daily activities of the patients [4,5].Despite the significant progress made in understanding the mechanism and therapeutic targets of AD, there is a lack of a definitive cure or effective treatment [6].Consequently, AD represents a growing societal challenge and an unmet medical need [1].
Several hypotheses have been proposed over the years to explain AD's pathogenesis, with the amyloid hypothesis being the prevailing paradigm [7,8].However, recent studies have questioned the validity of this hypothesis and suggested alternative explanations, including the tau hypothesis, chronic inflammation, and gut microbiota theories [9,10].Inflammation hypothesis, in particular, postulates that proinflammatory cytokine, tumour necrosis factor-alpha (TNFα), plays a crucial role in AD pathogenesis.TNFα is upregulated in the brains of individuals with AD and impairs cognitive function [11,12].Additionally, studies have shown that the modulation of TNFα leads to variations in amyloid plaque deposition, neuronal death, and cognitive deficits, which are hallmarks of AD [13,14].In general, there is compelling evidence to suggest that TNFα plays a significant role in the pathogenesis of AD [15].However, TNF inhibitors such as infliximab and etanercept do not cross the blood-brain barrier (BBB), which is a physical barrier that separates the brain from the peripheral circulation, limiting their efficacy in treating brain inflammation [16].Emerging evidence suggests that the plant Rauwolfia vomitoria (RV) possesses compounds capable of preventing neuronal damage and reducing inflammation in the brain with minimal side effects [17][18][19][20].This plant has exhibited promising therapeutic effects on cognitive deficit, among other plants that may have a beneficial effect on cognitive function [20][21][22].Rauwolfia vomitoria is an ethnomedicinal plant commonly used in traditional African medicine for various ailments, including inflammation [23][24][25].
Plant-derived compounds have been a focal point in drug discovery for centuries, and recent advances in computational chemistry and molecular modelling have expedited the process of identifying promising drug candidates from natural sources [26,27].In silico methods have been used to predict the biological activities of plant-derived compounds, thereby speeding up the process of identifying promising drug candidates at a reduced cost [28][29][30].This is particularly important for developing countries, where plant diversity is high and access to modern drug discovery technologies is limited.
Drug repurposing is the process of identifying new therapeutic uses for existing drugs.One approach to drug repurposing is based on the similarity of chemical structures between drugs.Furthermore, drug repurposing, particularly based on structural similarity, can potentially lead to the identification of new therapeutic uses for existing drugs.Based on the idea that if two molecules share similar structures, then they may have similar bioactivities [31][32][33][34].This approach is commonly used and aimed at identifying an analogue of an existing drug molecule that shares mechanisms of action with the original drug or compound [31,35,36].Therefore, in this study, we explored the potential of yohimbine, the most dominant compound in the stem bark of RV, as a TNFα binder and potential drug candidate.Additionally, we identified an existing drug metoserpate for TNFα inhibition based on structural similarities.

Study Workflow.
The study started with a systematic evaluation of the pharmacokinetic properties of yohimbine, the primary compound found in RV stem bark.In silico analysis using SwissADME was employed to comprehensively understand yohimbine's absorption, distribution, metabolism, and excretion (ADME) profile.Subsequently, molecular docking of yohimbine and the TNFα receptor was performed using SeeSAR software to determine their binding affinity.A structure similarity search for yohimbine was conducted to identify compounds with a similarity of at least 75%.The pharmacokinetic profile of the selected compound was evaluated using SwissADME.The most promising FDA-approved drug was chosen based on its pharmacokinetic properties and its ability to bind to the TNFα receptor.The selected drug underwent docking and molecular dynamics simulations using GROMACS software to assess its stability and potential in vivo performance.

2.2.
Retrieval and Preparation of 3D Protein Structure.The three-dimensional (3D) conformation of TNFα (PDB ID: 2AZ5; X-ray diffraction resolution: 2.10 Å), as previously reported by He et al. [37], was obtained from the Research Collaboratory for Structural Bioinformatics Protein Data Bank (RCSB PDB) [38] (https://www.rcsb.org).The retrieved protein structure was subjected to preparation using the Biovia Discovery Studio Visualizer v2021 [39].During protein preparation, all multiple chains were eliminated from the structure, resulting in the retention of chain "A" for subsequent molecular docking.Additionally, the water molecules and heteroatoms that were irrelevant to the investigation were removed during the protein preparation process.

Retrieval and Preparation of 3D Conformer Compounds.
The dominant compound, specifically yohimbine, in the stem bark of RV, was analysed in terms of their 3D conformer structure, as obtained from the PubChem database [40] (https://pubchem.ncbi.nlm.nih.gov/).The structure of yohimbine was processed using Avogadro v1.2.0 [41] (https://avogadro.cc) with the MMFF96 force field applied for the minimisation of the ligand after the addition of hydrogen atoms and the refinement of the geometry.BioMed Research International distribution, metabolism, and excretion) profile of the compounds was performed.Compliance with the Lipinski rule of 5 [42], which includes the parameters of molecular weight (MW), lipophilicity (log P), hydrogen bond acceptor (HBA), and hydrogen bond donor (HBD), was evaluated.Additionally, the GI absorption and penetration of the BBB of the compounds were examined using various models, Ghose's rule [43], Egan's rule [44], Muegge's rule [45], and Veber's rule [46].The Ghose rule defines acceptable compounds as having a molecular weight between 160 and 480 g/mol, a log P value between -0.4 and 5.6, the number of hydrogen bond donors less than or equal to 5, and the number of hydrogen bond acceptors less than or equal to 10. Egan's rule considers molecular weight, log P, the number of hydrogen bond donors, the number of hydrogen bond acceptors, and the number of rotatable bonds, while Veber's rule takes into account the number of rotatable bonds, the number of hydrogen bond donors, the number of hydrogen bond acceptors, and molecular weight.Muegge's rule assesses the acceptability of compounds based on their molecular weight, the number of hydrogen bond donors, the number of hydrogen bond acceptors, and the topological polar surface area (TPSA).The SwissADME open-access online tool was employed to evaluate the ADME profile of the compounds assessed in this study [47] (http://www.swissadme.ch).

Structural Similarity Search of DrugBank Compounds.
In this study, we used the DrugBank and SwissSimilarity tool [49] to investigate drug structural similarity using yohimbine structure as a query.Specifically, we used SwissSimilarity, which is an open-access web-based tool that allows molecular structure comparisons of drugs based on their chemical properties.The similarity search was performed against the DrugBank database [50], which provides comprehensive data on the chemical structure, pharmacology, and clinical applications of drugs.Notably, we opted for the 2D and 3D combined DrugBank option of the SwissSimilarity web platform for the search of structurally similar drugs, employing a similarity threshold of 75% and above.Equally, the chemical structure search feature was used for the yohimbine-centered approach to investigate drugs similar to yohimbine.

Pharmacokinetic Assessment of Identified DrugBank
Compounds.Pharmacokinetic assessment was carried out for identified drugs with structural similarity equal to or greater than 75% by evaluating the gastrointestinal (GI) absorption and BBB penetration for the selected drugs using the SwissADME web tool [47] (http://www.swissadme.ch) with focus on two important aspects (GI and BBB) of drug distribution in the body.

Results and Discussion
3.1.Pharmacokinetics of Yohimbine.The study of the pharmacokinetics of potentially therapeutic compounds is of clinical importance in the drug development process.
Elsewhere, about 40% of drug candidates do not pass the clinical trial stages [57] due to undesired absorption, distribution, metabolism, and excretion (ADME) profiles of the drug candidates.For a compound to be considered a good candidate depends on its exposure to the molecular target, which is determined by absorption and metabolism and particularly for central nervous system (CNS) drugs, an ability to cross the BBB [58].From Table 1, it can be inferred that yohimbine demonstrated high GI absorption and lipophilicity making it easier to cross the blood-brain barrier.
A few pharmacokinetic principles pioneered by Lipinski, Ghose, Veber, Egan, and Muegge were applied to assess the drug-likeness of the plant compound yohimbine.Yohimbine was subjected to Lipinski's rule of 5, per the rule; orally active drugs should not violate any of these four criteria: molecular weight ≤ 500, log P lipophilicity ≤ 5, number of hydrogen bond donors ≤ 5, and number of hydrogen bond acceptors ≤ 10 [42].Based on the physicochemical properties of yohimbine, none of the rules were violated (Table 1); this confers its use as an oral pharmaceutical drug.The total polar surface area (TPSA) for yohimbine was 65.56 Å 2 which is less than 140 Å 2 indicating good permeability in cellular lipid membranes according to Veber's rule [46].It is evident in literature that there is a strong correlation between high TPSA and low blood-brain penetration [59][60][61].The Ghose filter was applied to evaluate the drug-likeness of yohimbine; again, no rule was violated.Egan and Muegge's filters were employed to assess the oral bioavailability based on the physicochemical properties; once more, yohimbine was compliant with all the rules [62].

Molecular Docking of TNFα-Yohimbine.
The molecular docking result obtained between TNFα and yohimbine showed that there was a binding affinity Hyde score of -1.0 kJ/mol between the nitrogen atom at position 5 of the ligand and the amino acid residue Gln102 of TNFα 3 BioMed Research International (Figure 1(a) and Table 2).Additionally, there was another bond interaction (Hyde: 0.2 kJ/mol) between the oxygen atom at position 3 of the ligand and the amino acid residue Gln102.The observation of a binding affinity Hyde score of -1.0 kJ/mol between the nitrogen atom at position 5 of yohimbine and the amino acid residue Gln102 of TNFα suggests that yohimbine may bind to TNFα's active site and inhibit its proinflammatory effects.Additionally, the bond interaction between the oxygen atom at position 3 of yohimbine and Gln102 may contribute to the overall stability of the yohimbine-TNFα complex.Upon analysing the docking pose using Biovia Discovery Studio Visualizer, it was observed that the ligand formed two conventional hydrogen bond networks with the amino acid residue Gln102 of TNFα (Figure 1(b)).In addition to the conventional hydrogen bond networks, a nonconventional hydrogen bond network was also detected between the ligand and the amino acid residue Cys101 of TNFα.Furthermore, two pi-alkyl bond network interactions were observed between the ligand and the amino acid residue Arg103 of TNFα (Figure 1(b)).Hydrophobic interaction was also observed between yohim-bine and the TNFα residues Arg103 and Gln102.These results suggest that yohimbine has potential to bind to TNFα at its binding site and inhibit its proinflammatory effects.
From our study, yohimbine, an alkaloid with purported aphrodisiac properties and used for treating erectile dysfunction [63,64], has been identified as a potential inhibitor of TNFα, a cytokine that mediates inflammation in the central nervous system (CNS) and causes oxidative stress, apoptosis, and synaptic dysfunction in neurons [65].Neuroinflammation and resultant neurodegeneration can be precipitated by activated microglia, the resident immune cells of the CNS [66,67].Therefore, identifying small molecules capable of inhibiting TNFα could be therapeutically beneficial in treating neurodegenerative disorders associated with chronic inflammation.Here, we utilized SeeSAR, a structure-based drug design software tool, to study the interaction between yohimbine and TNFα [68,69].Our analysis has demonstrated that yohimbine exhibits a stable interaction with TNFα, as indicated by a Hyde score of -1.0 kJ/mol, suggesting favourable binding.Further examination of the molecular interactions has revealed key findings.Notably, a pi-alkyl  Additionally, a conventional hydrogen bond is observed between the hydrogen of the imine functional group of yohimbine and the amino acid residue glutamine at position 102 (Glu102) of TNFα.Furthermore, a nonconventional hydrogen bond network interaction between yohimbine and TNFα is observed at the amino acid residue position Cys101, which contributes significantly to the binding process.These molecular interactions, as illustrated in Figures 1(a) and 1(b), play a prominent role in driving the binding between yohimbine and TNFα.These findings provide valuable insights into the specific mechanisms underlying the interaction between yohimbine and TNFα, shedding light on the potential efficacy of yohimbine in modulating TNFα and its implications for addressing the pathogenesis of AD.Therefore, we posit that yohimbine may act as an inhibitor of TNFα and reduce its proinflammatory and neurotoxic effects in the CNS, which could explain the benefit reported for cognitive impairment and motor dysfunction [70,71].
Compared to other TNFα inhibitors like etanercept and infliximab, which are large molecules and have difficulty penetrating the blood-brain barrier and pose systemic safety concerns [72][73][74], yohimbine is a small molecule that can easily cross the blood-brain barrier and has a relatively good safety profile when used at low doses [75].Therefore, we propose yohimbine as a model molecule for the repurposing of an old FDA-approved drug (that may have superior bioavailability and safety profiles) for a new drug target (TNFα) inhibition, based on structure similarity search.

Identified Structurally Similar DrugBank Compounds.
The primary objective of this study is to explore structurebased drug design strategies in order to identify and repurpose known compounds, like yohimbine, for potential therapeutic use for the management of AD.To this end, we conducted a search of the DrugBank database for FDAapproved compounds that exhibited a high percentage structural similarity to yohimbine.From our analysis, a total of 10 compounds with a structural similarity of at least 75% to yohimbine were retrieved (Table 3).These compounds include metoserpate, deserpidine, 18-methoxycoronaridine, CP-320626, rescinnamine, reserpine, raubasine, methoserpidine, (7as,12ar,12bs)-1,2,3,4,7a,12,12a,12b-Octahydroindolo[2,3-a]Quinolizin-7(6h)-One, and vinburnine.After assessing the retrieved compounds for their current FDA approval status, four of the entries were found to have FDA approval.
Metoserpate (DB11530) demonstrated the highest percentage structure similarity (0.992%) to yohimbine, ability to traverse the BBB, high GI absorption, and preexisting approval for clinical use, thus making it an ideal candidate for further investigation (Tables 3 and 4).The observed high degree of similarity between metoserpate and yohimbine can be attributed to the presence of a pentacyclic yohimban skeleton, involving the formation of a carbocyclic ring from the C-17 to C-18 bond in a corynantheine precursor, as previously reported [63].

Pharmacokinetics of the Identified Structurally Similar
DrugBank Compounds.We assessed the GI absorption and the capacity to cross the BBB of the 10 compounds retrieved from the DrugBank database.Our findings showed that all 10 compounds had high GI absorption, indicating that they are likely to be well absorbed in the gastrointestinal tract (Table 4).However, only five of the compounds had the capacity to cross the BBB (Table 4), indicating that they may have potential therapeutic applications for the treatment of CNS disorders.These five compounds may be able to penetrate the BBB due to their physicochemical properties, such as their lipophilicity and molecular weight.
Further analysis revealed that out of the five compounds that are able to cross the BBB, only one (metoserpate) had FDA approval.Thus, metoserpate (DB11530) was the ideal candidate not only because it is the only FDA-approved drug, but also it exhibited high gastrointestinal absorption and a propensity to cross or permeate the blood-brain barrier.Metoserpate has a total polar surface area (TPSA) of 73.02 Å 2 contributing to its ability to permeate cellular membranes.It is evident in literature that TPSA values less than 73.02Å 2 are indicative of good permeability and satisfy Veber's rule [46].Metoserpate was thus selected for further analysis.

Molecular Docking of TNF-Alpha and Metoserpate. The binding affinity of TNFα and metoserpate was assessed using
Hyde's score method.This method seeks to address weak or questionable hydrogen bonds as well as indifferent scaffolds not contributing to the free energy in the protein-ligand complex [76,77].From Figure 2(a), it can be observed that the Hyde score was -1.1 kg/mol which confers a favourable interaction [77].The docking analysis revealed one hydrogen bond between the nitrogen atom at position 8 of metoserpate and the amino acid residue Gln102 of TNFα.These results suggest that the interaction between metoserpate and TNFα at this site may have potential therapeutic implications for the treatment of TNFα-related diseases (Figure 2(a) and Table 2).
When the docking simulation result was visualized using Biovia Discovery Studio Visualizer, one pi-alkyl bond network between metoserpate and TNFα binding site amino Gln102 = the key amino acid residue involved in the hydrogen bond formation.
5 BioMed Research International acid (AA) residue Arg103 and two salt bridge interactions between metoserpate and the binding site AA residue Glu104 of TNFα were observed.In addition, one conventional hydrogen bond network was observed between metoserpate and TNFα binding site residue Gln102 (Figure 2(b)).It is documented that conventional hydrogen bonds aid in the stability of complexes, hence conferring a good binding affinity [78,79].Consequently, the glutamic acid (Glu104) of the protein participated in two cation-pi interactions between the imine functional group and the benzene ring of metoserpate is shown in yellow.Cation-pi interactions play an important role in determining protein structure as well as contributing significantly to the binding energy of the complex formation [80].Arginine (Arg103) of the protein residue participated in a pi-alkyl interaction with the benzene ring of our target drug metoserpate.According to literature, pi-alkyl interactions have a greater propensity for stability when compared to alkyls bound to nonaromatic moieties in a ligand [81][82][83].
3.6.Molecular Dynamics Simulations 3.6.1.TNFα-Small Molecule and TNFα-Metoserpate.Numerous significant pharmaceuticals and hundreds of natural products with promising bioactivities contain indole alkaloids or have structures that are like indole alkaloids.Despite not always adhering to Lipinski's rules, such compounds frequently exhibit favourable pharmacokinetic profiles with respect to cyclic molecules.The values of the root mean square deviation (RMSD) affirm whether a close-match docked pose was predicted between the crystal and the predicted structures.It is evident in literature that an RMSD value ≤ 0 2 nm is fairly good [84][85][86].Figures 3(a) and 3(b) highlight the results of TNFα and the cocrystallized small molecule and TNFα and the target drug metoserpate both having their RMSD value ≤ 2 Å (0.2 nm) which confers a latent stable protein-ligand complex.
The RMSD between the TNFα-small molecule complex and the TNFα-metoserpate complex remained consistent throughout a 100 ns simulation.However, when comparing the TNFα-small molecule complex (Figure 3(a)) to the TNFα-metoserpate complex (Figure 3(b)), a more stable trajectory was observed in the TNFα-metoserpate complex.In the case of the TNFα-small molecule complex, it displayed stability from 20 ns to approximately 30 ns, followed by a deviation.It then regained stability until around 55 ns but experienced another deviation until 60 ns.From this point, it became stable again until approximately 75 ns, with  The observation of stable RMSD values throughout a 100 ns simulation suggests that the overall conformation of the TNFα-small molecule complex and TNFα-metoserpate complex remained relatively consistent during the simulation period [87].This stability is an important characteristic as it indicates that the complexes maintained their structural integrity and did not undergo significant conformational changes.TNFα-metoserpate complex exhibited a more stable trajectory compared to the TNFα-small molecule complex suggesting that the binding of metoserpate, a small compound, may have induced more favourable interactions and a more stable complex formation.This could be attributed to specific molecular interactions, such as hydrogen bonding, electrostatic interactions, or hydrophobic interactions between metoserpate and TNFα.These interactions Figure 2: (a) A 3D representation of the complex formed between TNFα and metoserpate using SeeSAR software.The Hyde score indicates the binding affinity of the ligand to the protein.The ligand's N8 atom form hydrogen bonds with the protein's Gln102 residue, contributing to the stability of the complex.(b) Molecular interactions of TNFα-metoserpate complex.3D visualization of the complex using Biovia Discovery Studio.The ligand (in grey, red, and purple) forms a hydrogen bond network with Gln102 through its N8 atom (in green dashed lines).The ligand also interacts with Arg103 through a pi-alkyl interaction (in mauve dashed line).A hydrophobic contact area (in light blue shade) is observed between the ligand and residues Gln102 and Glu104 (in light blue shade).Cation-pi interactions (in golden yellow) are formed between metoserpate and Glu104.BioMed Research International may contribute to a stronger binding affinity and a more stable conformation for the TNFα-metoserpate complex [88].
To describe the local conformational change in the TNFα and metoserpate and TNFα-small molecule complexes, the root mean square fluctuation (RMSF) was required.Figures 4(a) and 4(b) highlight the RMSF profile of the TNFα-small molecule and TNFα-metoserpate complexes, respectively.From the graph, stable fluctuations were observed with RMSF ≤ 0 2 nm in both instances [89].The TNFα-small molecule complex (Figure 4(a)) and TNFαmetoserpate complex (Figure 4(b)) both displayed reasonably low RMSF.However, the TNFα-metoserpate complex exhibited slightly higher fluctuations compared to the TNFα-small molecule complex.It is important to note that all the observed fluctuations in the TNFα-metoserpate complex were generally around 0.2 nm.On the other hand, in the TNFα-small molecule complex, fluctuations around atom positions 180 and 1520 were observed to be around 0.3 nm.
These fluctuations, measured in nanometers, indicate the degree of movement or flexibility of specific atoms within the complexes.The relatively low RMSF values suggest that overall, the complexes remained relatively stable during the simulation [90].However, the slightly higher fluctuations in the TNFα-metoserpate complex could imply that the binding of metoserpate induced some additional dynamics or flexibility in certain regions of the complex compared to the TNFα-small molecule complex [90].The specific atom positions 180 and 1520 in the TNFα-small molecule complex experienced slightly higher fluctuations around 0.3 nm.These positions could correspond to specific residues or functional regions within the complex.The increased fluctuation at these positions may indicate potential conformational changes or greater flexibility in those regions, possibly influenced by the presence of the small compound or specific interactions between the compound and TNFα [90].
The radius of gyration (Rg) monitors the compactness of the protein structure coupled with the binding patterns of the drug and protein in direct relation to the folding rate [91].A conformational change occurs when a ligand or lead molecule attaches to the protein, changing the radius of gyration [92].The TNFα-small molecule complex (Figure 5(a)) and TNFα-metoserpate complex (Figure 5(b)) exhibited similar total radius of gyration values, both measuring approximately 1.52 nm.A smaller radius of gyration indicates a more compact and tightly packed structure, while a larger radius of gyration suggests a more extended or flexible conformation [90].The fact that both the TNFαsmall molecule complex and TNFα-metoserpate complex demonstrated a total radius of gyration around 1.52 nm suggests that they possess comparable overall compactness, indicating a compact and stable conformation [93].This similarity in size could indicate that the binding of both the small molecule and metoserpate did not significantly alter the overall conformation or compactness of the TNFα complex.
3.6.2.Bond Network Evaluation of Metoserpate and TNF-Alpha Complex following Molecular Dynamics Simulation.The post-MD simulation analysis revealed significant changes in the metoserpate-TNFα complex compared to the pre-MD simulation complex.Our findings demonstrated that metoserpate established multiple bond network interactions with the AAs in the binding site of TNFα.Specifically, a conventional hydrogen bond (cH-bond) was formed between the oxygen of the carboxylic acid methyl ester of metoserpate and the amino acid residue Lys98 of TNFα.Conventional hydrogen bonds are known for their strength and contribute to strong binding affinity.Additionally, several nonconventional hydrogen bonds (ncH-bonds) were observed between metoserpate and the AAs Ser99, Glu104, Pro113, Tyr115, and Glu116.Metoserpate also engaged in a pi-alkyl interaction with Tyr115 and Pro117, as well as two cation-pi interactions with Glu104 and Glu116.These interactions played a crucial role in the stability and specificity of the complex (Figure 6).Hydrophobic contact area was also established between metoserpate and the binding site AA residues Lys98 and Tyr115.
In contrast, the bond network analysis conducted prior to the MD simulation revealed specific interactions between metoserpate and TNFα, including a cH-bond network with Gln102, a pi-alkyl interaction with Arg103, and cation-pi interactions with Glu104.However, the subsequent simulation analysis yielded intriguing findings, indicating an enhanced binding affinity and selectivity of metoserpate towards TNFα.This improvement in binding was accompanied by the generation of more favourable and specific interactions.
These results are further supported by the observed flexibility in the root mean square fluctuation (RMSF) output of the TNFα and metoserpate complex.The MD simulations have provided valuable insights into the intricate molecular interactions between the TNFα and metoserpate, unravelling the complexities of protein-ligand complexes.
By elucidating the dynamic behaviour and uncovering the structural changes that occur during the simulation, the MD simulations offer a deeper understanding of the binding mechanism and contribute to the overall comprehension of the interactions between TNFα and metoserpate.
3.6.3.Overall Bond Network Assessment.The Hyde scoring method has proven to be a valuable computational tool in drug discovery for estimating the binding affinity between a protein and a ligand, utilizing their interaction energy [94].In the present study, we employed the Hyde score assessment method to evaluate the binding affinity of two ligands, metoserpate and yohimbine, with the protein TNFα,    BioMed Research International with a specific focus on the amino acid residues within the binding site (Table 2).
The study's findings revealed that metoserpate exhibited slightly lower Hyde score (-1.1 kJ/mol) in comparison to yohimbine (-1.0 kJ/mol and 0.2 kJ/mol) when interacting with Gln102 (Table 2).This indicates that metoserpate possesses a marginally better binding affinity with TNFα when compared to yohimbine, although the difference observed is relatively small.These results shed light on the relative strengths of the interactions between metoserpate and TNFα, providing insights into the binding affinity.This information contributes to the understanding of the potential efficacy of metoserpate as a potential therapeutic agent targeting TNFα in the context of AD management.
The post-MD simulation analysis revealed the involvement of amino acid Lys98 in the conventional hydrogen bond formation, as well as the formation of a nonconventional hydrogen bond network with Pro113, which was originally not part of the binding site AA residues (Figure 6).This post-MD simulation analysis generated more bond diversity, and bond number compared to the TNFαyohimbine and TNFα-metoserpate complexes.This demonstrates the importance of post-MD simulation analysis in providing a more comprehensive understanding of proteinligand interactions beyond what can be predicted through initial scoring methods alone.
The findings of this study also suggest that the binding affinity of a ligand with a protein may be influenced by amino acid residues outside of the initial binding site.This is consistent with previous studies that have shown the importance of protein flexibility and dynamics in ligand binding [95].It is possible that the nonconventional hydrogen bond network identified in the post-MD simulation analysis plays a critical role in the binding affinity between TNFα and metoserpate.

Conclusion
In summary, our study employed the Hyde score assessment method to evaluate the binding affinity of metoserpate and yohimbine with TNFα, with a specific focus on the binding site amino acid residues.While metoserpate generated a lower Hyde score than yohimbine with the key binding site amino acid Gln102, further investigation using postmolecular dynamics (MD) simulation analysis demonstrated the involvement of additional amino acid residues in the binding affinity.The results indicated that metoserpate has the potential to inhibit TNFα and thus presents as a promising candidate for further study as a therapeutic agent for TNFα-related diseases.Additionally, our work showcases the utility of yohimbine as a query compound to identify structurally similar drugs from the DrugBank database in the context of drug repurposing.Specifically, our study identified metoserpate as a potential inhibitor of TNFα using a computational approach that combined molecular docking and MD simulation.This approach allowed for a more comprehensive and nuanced understanding of the binding affinity of metoserpate with TNFα and provided insights into the potential mechanisms of inhibition.Furthermore, our use of yohimbine as a query compound helped identify metoserpate as a structurally similar compound with potential therapeutic properties.Overall, these findings represent a significant step forward in the development of metoserpate as a potential therapeutic agent for TNFα-related diseases.However, further research is needed to validate these findings through in vitro and in vivo (in a physiologically relevant cell line, fly models, and/or animal models) studies and to optimize the efficacy of metoserpate as a drug candidate.

2. 4 .
Pharmacokinetic Assessment of Yohimbine.In this study, a comprehensive analysis of the ADME (absorption, 2

Figure 1 :
Figure 1: (a) A 3D representation of the complex formed between TNFα and yohimbine using SeeSAR software.The Hyde score indicates the binding affinity of the ligand to the protein.The ligand's N5 and O3 atoms form hydrogen bonds with the protein's Gln102 residue, contributing to the stability of the complex.(b) Molecular docking of yohimbine with TNFα protein.3Dstructure of the protein-ligand complex visualized using Biovia Discovery Studio.The ligand (in grey, red, and purple) forms a hydrogen bond network with Gln102 (in green dashed lines) through its N5 and O3 atoms and a nonconventional hydrogen bond network between the O3 atom of yohimbine and Cys101 (in light green dashed lines).The ligand also interacts with Arg103 through a pi-alkyl interaction (mauve dashed line).The ligand is surrounded by a hydrophobic contact area (in light blue shade) involving Gln102 and Arg103.

Figure 3 :
Figure 3: Trajectories of the overall RMSD: (a) TNFα-small molecule complex; (b) TNFα-metoserpate complex.RMSD of the various complexes with respect to the starting structure over 100 ns MD simulation.The x-axis represents the simulation time in nanoseconds.The y-axis represents RMSD in nanometers.

Figure 4 :
Figure 4: Residue-wise RMSF profiles of the TNFα and various ligand complexes: (a) TNFα-small molecule complex; (b) TNFαmetoserpate complex.The x-axis represents the atom number.The y-axis represents RMSF in nanometers.

Figure 5 :
Figure 5: Radius of gyration profiles of the TNFα and various ligand complexes: (a) TNFα-small molecule complex; (b) TNFα-metoserpate complex.The x-axis represents the time in picoseconds.The y-axis represents Rg in nanometers.

Figure 6 :
Figure 6: Molecular interactions between TNFα and metoserpate after molecular dynamics simulation.The protein residues involved in binding are represented in disc shape with different colours.The ligand is shown in ball-and-stick representation with different colours for different atom types.The molecular interactions are depicted by dashed lines with different colours indicating different types of interactions: conventional hydrogen bonds (deep green), nonconventional hydrogen bonds (light green), pi-alkyl interactions (mauve), and cation-pi interactions (golden yellow).

Table 2 :
Hyde's score estimates and TNFα binding site amino acid residues.

Table 3 :
Drugs that are 75% or more structurally similar to yohimbine.