In silico molecular targets, docking, dynamics simulation and physiologically based pharmacokinetics modeling of oritavancin

Abstract

Introduction

Oritavancin is a semi-synthetic lipoglycopeptide antibiotic primarily used to treat serious infections caused by Gram-positive bacteria. The aim of this study was to elucidate possible molecular targets of oritavancin in human and microbes in relevance to its mechanism of action and model its pharmacokinetics for optimal dose selection in clinical practice.

Methods

Computational methods were used in this study which include target prediction, molecular docking, molecular dynamics simulation, pharmacokinetics prediction, and physiological-based pharmacokinetics (PBPK) modeling.

Results

Oritavancin was moderately soluble in water and did not permeate the blood-brain barrier. Seven molecular targets were identified in humans. Molecular docking results showed highest binding affinity of oritavancin with PI3-kinase p110-gamma subunit (−10.34 kcal/mol), followed by Acyl-CoA desaturase (−10.07 kcal/mol) and Cytochrome P450 2C19 (−8.384 kcal/mol). Oritavancin PBPK modelling in adult human showed that infusion has lower peak concentrations (Cmax) compared to bolus administration, with 1200 mg dose yielded Cmax of 16.559 mg/L, 800 mg dose yielded Cmax of 11.258 mg/L, and 200 mg over 3 days dose yielded Cmax of 7.526 mg/L. Notably, infusion gave extended half-life (t_1/2) for all doses and slightly higher clearance rates compared to bolus, particularly for the 1200 mg and 800 mg doses. The results corroborated existing clinical pharmacokinetic data, and confirmed the model’s accuracy and predictive capability.

Conclusion

This comprehensive computational study has provided invaluable insights into the pharmacological profile of Oritavancin, aiding its further development and optimization for clinical use.

Introduction

Antibiotics are the most significant class of pharmaceuticals that has helped human to fight against bacteria and saved millions of lives. However, multidrug-resistant (MDR) bacteria that causes antimicrobial resistance (AMR), are increasing noticed across the world and responsible for deaths of about 4.5 million people in 2019. This presents a significant challenge to human race and 20 million deaths have been forecasted for 2050 [1, 2]. Therefore, there is urgent need to discover novel compounds and optimize existing antibiotics in order to salvage the ongoing crisis. The ongoing efforts in the discovery of antibiotics has found oritavancin to be one of the effective long-acting antibiotics of the 21st century, but presently limited data are available on its molecular targets in mammals.

Oritavancin (Fig. 1) is a second-generation semisynthetic lipoglycopeptide antibiotic obtained by modifications of Chloroeremomycin, a naturally occurring glycopeptide isolated from Kibdelosporangium orienticin [3]. Oritavancin is active against gram-positive aerobic bacteria such as enterococci, staphylococci, streptococci, and anaerobic bacteria such as Clostridium difficile, C. perfringens, Peptostreptococcus spp., and Propionibacterium acnes [4,5,6]. Oritavancin has been used for the treatment of acute bacterial skin and skin structure infections (ABSSSIs) potentially caused by susceptible isolates of some gram-positive microorganisms that include methicillin-resistant Staphylococcus aureus (MRSA) [3, 7,8,9,10].

Oritavancin is structurally similar to vancomycin and has similar spectrum of activity but with lower minimum inhibitory concentrations [11], which make it a valuable option in the treatment of serious bacterial infections. Oritavancin has a multifaceted mechanism of action that exhibited concentration-dependent effects against gram-positive organisms [12, 13]. It works against susceptible gram-positive organisms via three separate mechanisms: (i) it binds to the stem peptide of peptidoglycan precursors, D-alanyl-D-alanine terminus of lipid II, inhibiting transglycosylation (polymerization), which normally occurs during cell wall synthesis; (ii) it inhibits crosslinking during bacterial cell wall biosynthesis via binding to cell wall pentaglycyl peptide bridging segments; and (iii) it disrupts the bacterial cell membrane by interfering with membrane integrity, which eventually leads to cell death by cascade of biosignaling mechanisms [3, 14, 15]. The cell wall is vital for the survival and replication of bacteria, making it a primary target for antibiotic therapy [16].

In this study, computational approaches were utilized in order to have holistic comprehension of actions of oritavancin in biological systems. The approach used include: (i) Target prediction, which involves identifying potential biological targets that a drug molecule might interact with, which can provide insights into its mechanism of action and possible therapeutic effects. This is achieved using computational tools that analyze the chemical structure of the compound and compare it with known databases of biological targets; (ii) Prediction of Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) profiles of a drug. This computational method predicted how the drug will behave in the body, including its bioavailability, potential metabolic pathways, excretion routes, and possible toxicological effects, which are crucial for understanding drug safety and efficacy [17]; (iii) Molecular docking, which is a technique that predicts the preferred orientation of a drug molecule when bound to its target protein. This method simulated the interaction between the drug and the target, providied insights into the binding affinity which is essential for drug design and optimization [18, 19]; (iv) Molecular dynamics (MD) simulation, which involves computationally simulating the physical movements of atoms and molecules over time. This method provides a detailed presentation of structural dynamics and conformational changes of the drug-target complex, and helped to understand the stability and behavior of the interaction under mimicked physiological conditions [20, 21], and (v) Physiologically-based pharmacokinetic (PBPK) modeling, utilizes mathematical models to predict the ADMET of a drug in human or animal such as rat. PBPK models integrate physiological parameters, such as organ sizes and blood flow rates, with drug-specific properties to simulate concentration-time profiles in various tissues and organs [22, 23]. In this study, holistic integration of these approaches helped in the prediction of adjuvant target and pharmacokinetics modelling of oritavancin in human.

The aim of this study was to elucidate possible molecular targets of oritavancin in human and microbes in relevance to its mechanism of action and model its pharmacokinetics for optimal dose selection in clinical practice.

Fig. 1

Chemical structure of oritavancin, adapted from Rosenthal et al. [9]

Materials and methods

In silico target prediction

The chemical structure of Oritavancin (PubChem CID: 16136912) was obtained from NCBI PubChem Compound database (https://pubchem.ncbi.nlm.nih.gov/) in SMILES formats. The SMILES was used for target prediction analysis on the SuperPred webserver (https://prediction.charite.de) and targets related to bacterial infection were selected. Also, the curated targets of oritavancin were obtained from DrugBank database (https://go.drugbank.com/).

Molecular docking studies

The molecular docking studies were carried out as previously reported [19, 20]. Briefly, the three-dimensional AlphaFold structure of the predicted target proteins were obtained from the Uniprot database (www.uniprot.org) and Protein databank (https://www.rcsb.org/) in pdb format. The structure of Oritavancin (ligand) SMILES converted to chemical structure and subjected to 3D structure optimization using ACDLab/Chemsketch software, and saved in.mol format. PyMol v3.0 was used for ligand file conversion from.mol to.pdb and for the preparation of protein chain A with removal of water and existing ligands. Prior to docking, the ligands and target proteins structures in pdb format were formatted to pdbqt by using AutoDock Tools (ADT) v1.5.6 [24]. Ligand-protein molecular docking was carried out using AutoDock Vina v1.2.3 [25, 26]. Afterward, binding affinity was obtained and close interactions of binding of the ligands with the targets were analyzed and visualized using PLIP webserver at https://plip-tool.biotec.tu-dresden.de/plip-web/plip/index [27] and PyMol v3.0 (Schrodinger LLC).

Molecular dynamics simulation

Two ligand-protein complexes with best binding affinity were used for this analysis. Maestro’s protein preparation wizard was used preprocessing. Desmond software by Schrödinger LLC v2021-1 was used to implemented the system simulation of 100 ns [20, 28, 29]. The system setup includes OPLS-2005 force field, orthorhombic box with TIP3P water model containing 0.15 M NaCl counter ions was used to established physiological conditions, and 1 atm pressure and 300 K temperature were used for the simulation. The models underwent energy minimization before simulation. During full system simulation, the trajectories were saved at every 100ps. The post-simulation trajectories analysis was conducted to determine (i) root-mean-square deviation (RMSD) which measures the average distance between the atoms of a reference structure and those of a trajectory ensemble, (ii) root-mean-square fluctuation (RMSF) which quantifies the flexibility or mobility of individual atoms or residues within a biomolecular system throughout the simulation, and (iii) protein-ligand contact profile which shows interactions such as hydrogen bonds, hydrophobic and electrostatic interactions.

Physicochemical parameters preparation

Physicochemical parameters of oritavancin were predicted as described by Shen et al. [23]. The SMILES of oritavancin was used to queried pkCSM server (http://biosig.unimelb.edu.au/pkcsm/). Also, information on oritavancin was obtained from the Drugbank (https://go.drugbank.com/).

PBPK model specification and simulation

Simple PBPK modeling was conducted on Teoreler web-based application (www.teoreler.com), which was developed by the Department of Pharmacology and Therapeutics, University of Liverpool, UK. Adult human single drug model was used for this study. ChatGPT v4 from OpenAI was used to specify globally consensus values for physiological parameters in the model development [30, 31], and doses were based on existing literature [4, 7, 8]. Intravenous bolus and intravenous infusion were simulated in this study.

Results

In this study, oritavancin was investigated using comprehensive computational methods. The molecular target results on SuperPred webserver showed three human molecular targets (Acyl-CoA desaturase, PI3-kinase p110-gamma subunit and P-glycoprotein 1), with over 90% accuracy in relation to bacterial infection (Table 1) and four human molecular targets (Cytochrome P450 (CYP) 3A4, CYP2D6, CYP2C9 and CYP2C19), were obtained from DrugBank database (Table 2). Molecular docking result (Table 3) showed that Oritavancin exhibited a high binding affinity for PI3-kinase p110-gamma subunit with a docking score of −10.34 kcal/mol, followed by Acyl-CoA desaturase (−10.07 kcal/mol) and Cytochrome P450 2C19 (−8.384 kcal/mol). The binding pose of the interaction of oritavancin with these proteins are presented in Figs. 2 and 3.

The results of the ligand-protein complexes MD simulations showed stability of the binding interaction of oritavancin with PI3-kinase p110-gamma subunit (P48736) and acyl-CoA desaturase (Q86SK9) respectively as presented in Fig. 4. For the oritavancin- P48736 complex, the RMSD of the protein was 9.0 Å and the ligand was 10.5 Å over the 0–100 ns simulation period (Fig. 4a). RMSF analysis showed maximal fluctuation at amino acid residues 1–80 on the N-terminal (Fig. 4b). Ligand-protein interactions involved amino acid residues such as ASN299, GLU301, LEU823, SER824, LYS883, and ASP884, contributing to hydrogen bonds, ionic interactions, hydrophobic interactions, and water bridges (Fig. 4c). Similarly, for the oritavancin-Q86SK9 complex, the RMSD of the protein was 6.2 Å and the ligand was 13.5 Å over the 0–100 ns simulation period (Fig. 4d). RMSF analysis indicated maximal fluctuation at amino acid residues 1–35 at the N-terminal (Fig. 4e). Ligand-protein interactions involved amino acid residues such as HIS57, ALA60, LEU226, THR229, ASN233, TYR245, and LEU264, contributing to various interactions (Fig. 4f). Detailed ligand-protein atom interactions were presented in Fig. 5.

The ADMET prediction results of oritavancin (Table 4) showed that it exhibited poor absorption characteristics; low water solubility (log mol/L of −2.892), Caco-2 permeability (log Papp of −1.874), 0% intestinal absorption in humans, and low skin permeability (log Kp of −2.735). As a substrate of P-glycoprotein, oritavancin is actively efflux from cells, and it does not inhibit P-glycoprotein I or II. In terms of distribution, oritavancin has a volume of distribution (VDss) of −0.274 log L/kg. This indicate that oritavancin is largely confined to the plasma, not widely distributed in tissues. With 34% unbound in human plasma, oritavancin has a moderate fraction unbound. It has very poor blood-brain barrier (BBB) permeability (log BB of −3.923), and it has extremely low CNS permeability (log PS of −6.421). Oritavancin is metabolized by CYP3A4 but it’s not a substrate of CYP2D6 and does not inhibit CYP1A2, CYP2C19, CYP2C9, CYP2D6, or CYP3A4. This suggest that oritavancin has minimal potential for drug-drug interactions via CYP inhibition. The total clearance is low (log ml/min/kg of −2.059), and it is not a substrate of the renal OCT2 transporter. Toxicity predictions show that oritavancin is not mutagenic according to the AMES test, and it has a maximum tolerated dose in humans of 0.438 log mg/kg/day. It does not inhibit hERG I or II, which indicate a low risk of cardiotoxicity related to QT prolongation. Oritavancin’s LD₅₀ in rats is 2.482 mol/kg, and its chronic toxicity (LOAEL) is 11.816 log mg/kg_bw/day. Oritavancin was found to be non-hepatotoxic, cause no skin sensitization, and has moderate environmental toxicity with a T. pyriformis toxicity of 0.285 log µg/L and minnow toxicity of 12.298 log mM.

Parameters used for PBPK modelling were presented in Table 5, and the pharmacokinetics summary results were presented in Table 6. The simulated plasma concentration (log-scale) vs. time (hour) profiles were shown in Fig. 6. PBPK results for Oritavancin IV bolus showed that higher doses gave proportional higher peak plasma levels, with 1200 mg has 35.497 mg/L, 800 mg has 23.747 mg/L, and 200 mg over 3 days showed 10.97 mg/L. Trough levels remain low across all doses, indicating a rapid decline in plasma concentration post-peak. The Area Under the Curve (AUC) values reflect the total drug exposure over time, showing a direct relationship with dose: 3751.9 mg×h/L for 1200 mg, 2489.6 mg×h/L for 800 mg, and 1885.3 mg×h/L for 200 mg. The half-life varies significantly, with the 200 mg dose has an unusual extended half-life of 80.803 h which likely due to the cumulative effect over three days. Considering Tmax (Time to Reach Cmax), both 1200 mg and 800 mg doses reach Cmax rapidly (0.1 h), whereas the 200 mg dose showed an anomalous Tmax. Vss (Volume of Distribution at Steady State) indicates a moderate distribution into body tissues, with values ranging from 0.5255 to 0.8369 L/kg. Clearance rates are consistent across doses, around 6 L/h, which suggest dose-independent clearance mechanisms.

PBPK results for Oritavancin IV infusion showed lower peak concentrations compared to bolus administration, with 1200 mg yielding 16.559 mg/L, 800 mg at 11.258 mg/L, and 200 mg over 3 days at 7.526 mg/L. Slightly higher trough levels than bolus administration, but still low. Slightly lower than bolus administration, but followed a similar dose-response relationship. There are notable extended half-lives for all doses, which indicate prolonged drug presence in the system. Tmax reflects the slow infusion process of about 12 h for 1200 mg and 800 mg doses, and 57.373 h for the 200 mg dose. It showed slightly higher clearance rates compared to bolus, particularly for the 1200 mg and 800 mg doses. Furthermore, the results showed that adipose, bone and brain have the lowest concentrations while liver, lungs, spleen, and kidney have the highest concentrations, for both infusion and IV bolus at 200 for 3 days doses as well as 800 mg and 1200 mg single doses.

Table 1 Predicted molecular targets of oritavancin from SuperPred

Table 2 Molecular targets of oritavancin from DrugBank

Table 3 Molecular docking result

Fig. 2

2D interaction of Oritavancin and the molecular targets A Cytochrome P450 3A4. B Cytochrome P450 2D6. C Cytochrome P450 2C9. D Cytochrome P450 2C19. E PI3-kinase p110-gamma subunit. F Acyl-CoA desaturase. G P-glycoprotein 1

Fig. 3

3D interaction of Oritavancin and the molecular targets A Cytochrome P450 3A4. B Cytochrome P450 2D6. C Cytochrome P450 2C9. D Cytochrome P450 2C19. E PI3-kinase p110-gamma subunit. F Acyl-CoA desaturase. G P-glycoprotein 1

Fig. 4

MD simulation results A RMSD of Oritavancin and P48736. B RMSF of P48736 on binding to Oritavancin. C Oritavancin and P48736 contact profile. D RMSD of Oritavancin and Q86SK9. E RMSF of Q86SK9 on binding to Oritavancin. F Oritavancin and Q86SK9 contact profile

Fig. 5

A representation of detailed interactions of A Oritavancin and P48736. B Oritavancin and Q86SK9

Table 4 ADMET prediction results

Table 5 Parameters used for oritavancin PBPK modelling

Fig. 6

Time-concentration pharmacokinetic profile

Table 6 Pharmacokinetic summary of oritavancin

Discussion

This study leverages in silico methodologies to predict molecular targets, perform docking studies, and develop a physiologically based pharmacokinetics (PBPK) model for oritavancin, an antimicrobial agent used for the treatment of gram-positive infections. The integrative approach taken in this research provides several noteworthy insights into the pharmacological and pharmacokinetics profiles of oritavancin.

In this study, seven molecular targets were obtained for oritavancin which include potential host-targeted proteins (Acyl CoA desaturase and PI3-kinase p110-gamma subunit) and metabolizing enzymes (CYP3A4, CYP2D6, CYP2C9, and CYP2C19), and efflux transporter (P-glycoprotein 1). P-glycoprotein (P-gp) efflux action and cytochrome P450s (CYPs) enzyme activity, have significant impact on the drug pharmacokinetics by clinically altering the administered drug efficacy or resulting to various adverse side-effects due to drug–drug interactions (DDIs), as in the case of multi-administration of drugs [33]. Low GIA of oritavancin could be attributed with its heavy molecular weight and not first-pass metabolism.

Host-targeted proteins aid in the interaction between a microbe and the human body. They can either facilitate or hinder the ability of a pathogen to infect human cells. Acyl-CoA desaturases are enzymes that introduce double bonds into fatty acyl-CoA substrates (desaturation of fatty acids), playing a crucial role in fatty acid metabolism. Both humans and bacteria possess Acyl-CoA desaturases (ACD), but their structure, function, and regulation differ due to their distinct metabolic roles; for instance, human ACD is only endoplasmic reticulum membrane bound and act for fatty acid biosynthesis and membrane fluidity while bacteria ACD is both membrane-bound and soluble, and adapts membrane fluidity to environmental changes such as temperature and other stress [34, 35]. Bacteria rely on the synthesis of specific lipids for the integrity and function of their cell membranes. Inhibiting ACD could disrupt the production of these essential fatty acids, compromising the structural and functional integrity of bacterial membranes. This disruption of fatty acid synthesis has been shown to affect bacterial viability and resistance mechanisms [36]. By interfering with lipid metabolism, an antimicrobial agent targeting ACD may lead to weakened bacterial defenses and increased susceptibility to other antimicrobial agents. Studies have indicated that targeting lipid biosynthesis pathways can be an effective strategy in developing new antimicrobials [37].

Phosphoinositide 3-kinase (PI3K) is a family of enzymes involved in cellular functions such as growth, proliferation, differentiation, motility, survival, and intracellular trafficking. The PI3K family is divided into different classes, with Class I PI3Ks playing a key role in cellular signaling in both humans and bacteria. PI3K p110-gamma (PIK3CG) is the catalytic subunit of Class I PI3Ks and it play a role in the activation and migration of neutrophils, macrophages and other immune cells to sites of infection [38]. PI3K signaling pathway has been reported to be essential and sufficiently involved in cellular entry of Pseudomonas aeruginosa [39], and it has been proposed be an effective adjuvant target for the treatment of bacterial infections [40]. Modulating P13K activity can enhance the host’s immune response to bacterial infections. It has been shown that some bacteria can manipulate host cell signaling pathways, including P13K pathways, to evade the immune response and establish infections [41]. Inhibiting P13K p110-gamma could disrupt these bacterial strategies, indirectly enhancing antibacterial activity by preventing bacteria from hijacking host cell signaling mechanism. P13K p110-gamma has also been shown to exhibit anti-inflammatory properties which could be beneficial in reducing excessive inflammation during bacterial infections [42]. The identification of P13K p110-gamma suggests that oritavancin might influence the host’s immune response to infections.

Molecular docking was used to predict the preferred orientation of the ligand when bound to the target protein to form a stable complex. Binding affinity have biological implications on drug efficacy and potency, the lower (more negative) the binding energy, the stronger the interaction [43]. In this study, the highest binding affinity was observed for the P13-kinase p110-gamma subunit and Acyl CoA desaturase, with the binding energies of −10.34 kcal.mol⁻¹ and −10.07 kcal.mol⁻¹ respectively. Low binding affinity ( ≤ −5.00 kcal.mol⁻¹) indicates strong binding between the ligand and the target, suggesting that the ligand is likely to be a potent inhibitor or inducer of the target [44].

Furthermore, MD simulations was used to assessed the binding affinity and stability of a protein-ligand complex, using various metrics which include RMSD, and RMSF. In this study, RMSD obtained were above 2.0 Å. RMSD measures the average deviation of a set of atomic positions (typically the backbone or all heavy atoms of the protein) from a reference structure over time. RMSD of about 2.0 Å indicates that the proteins had undergone relatively small conformational changes and were, thus, stable during the simulation [21]. This dynamic perspective is instrumental in understanding the stability and functional implications of ligand-protein interactions, guiding the refinement of molecular models and the design of more effective therapeutics. Monitoring RMSD over time helps in identifying significant conformational changes, and large deviations may indicate flexibility or instability [43]. RMSF provides insights into the flexibility of individual residues or regions within the protein [19].

The results showed that oritavancin is poorly absorbed from the gastrointestinal tract, thus intravenous administration is inevitable to achieve therapeutic systemic concentrations. P-glycoprotein is a membrane-bound efflux transporter [33], and the efflux mechanism can reduce the intracellular concentration of antibiotics, thereby contribute to antimicrobial resistance. Bacterial cells can express efflux pumps analogous to P-gp, which help them evade the effects of antibiotics by pumping them out of the cells, thus decreasing the drug’s efficacy [45]. P-gp can influence the pharmacokinetic profile of antimicrobial agents, by lowering therapeutic concentration in target tissues [46]. By potentially inhibiting P-gp, oritavancin could enhance its own intracellular concentration and that of co-administered antibiotics, thereby improves antimicrobial efficacy. Cytochrome P450 enzymes including CYP34A, CYP2D6, CYP2C9, and CYP2C19, play a significant role in determining the pharmacokinetic profiles of drugs by affecting their absorption, distribution, metabolism and excretion (ADME). Antibiotics can act as inhibitors or inducers of P450 enzymes, leading to significant drug-drug interactions. For instance, certain antibiotics can inhibit CYP34A, potentially increasing the plasma concentration of co-administered drugs that are CYP34A substrates [47]. It has been shown that some bacteria could induce CYP34A expression in host cells, potentially increasing the metabolism and clearance of antibiotics. This induction might contribute to antibiotic resistance by reducing drug concentration below therapeutic levels [48]. Oritavancin has been reported to be non-nephrotoxic, non-hepatotoxic, non-ototoxicity, and non-QT prolongation; however, adverse reactions associated with oritavancin in clinical trials included nausea, vomiting, headache, diarrhea, phlebitis, extravasation, infusion-related reactions, hypersensitivity, cellulitis, and constipation [9].

The ADME parameters used in drug discovery to identify acceptable drug candidates have a general range [49, 50]. Typically, solubility should be within 1–1000 µM in aqueous solution at physiological pH (pH 7.4) to ensure adequate absorption and bioavailability. High permeability is indicated by Caco-2 cell permeability (Papp) values greater than 10 e-6 cm/s and effective permeability coefficient (Peff) in the PAMPA assay should be above 1 e-6 cm/s. Optimal LogP values range from 1 to 3. A value within 0 to 5 is generally acceptable, with a preference for values around 2–3 to balance solubility and permeability. Ideally, the plasma protein binding should be less than 95% to ensure sufficient free drug concentration in the bloodstream. A half-life (t_1/2) of 4–12 h is considered optimal for maintaining effective plasma concentrations without frequent dosing. Acceptable clearance (CL) rates are generally between 10 and 20 mL/min/kg in humans. Lower clearance rates are preferable to maintain adequate plasma concentrations. An optimal volume of distribution (Vd) ranges from 0.7 to 1 L/kg, indicating balanced distribution within the body. Good metabolic stability is indicated by a half-life greater than 30 min in human liver microsomes, suggesting slower metabolism and better bioavailability. Also, oral bioavailability should ideally be greater than 30% to ensure sufficient systemic exposure after oral administration. Overall, oritavancin presents challenges in achieving effective bioavailability and CNS targeting due to poor absorption and distribution characteristics. However, it has a low risk of toxicity and drug interactions, making it a potential candidate for further development with optimization.

Machine learning and artificial intelligence (AI) tools have been pivotal in advancing drug discovery and useful in addressing potential limitation in PBPK modeling [51,52,53]. ChatGPT optimizes research processes by rapidly parsing vast amounts of literature and identifying key findings, thereby saves considerable literature review time for researchers, and facilitating the exploration of complex scientific problems [31]. AI-assisted PBPK approach can serve as an alternative non-animal method to facilitate nanomedicine research and development. The PBPK modeling provided a comprehensive simulation of oritavancin’s pharmacokinetics within the human body.

Intravenous bolus and intravenous infusion were simulated in this study. Both methods are important in medical settings depending on the urgency of treatment and the characteristics of the medication being administered. The main difference between an intravenous (IV) bolus and an infusion lies in how the medication or fluid is administered: (i) Speed—bolus is rapid (short duration), whereas infusion is slow and continuous (long duration); (ii) Administration—bolus is a single, quick injection, while infusion is a continuous delivery over time; and (iii) Purpose—bolus is used for immediate effect or when rapid delivery is needed, while infusion is used for sustained delivery or maintenance.

In this study, higher Cmax values in bolus administration suggest more rapid and higher peak concentrations, potentially enhancing immediate therapeutic effects but also increasing the risk of concentration-dependent toxicity. Ctrough represents the lowest concentration reached before the next dose. This is crucial for understanding the drug’s efficacy over time and ensuring concentrations remain above the minimum inhibitory concentration (MIC) to maintain antimicrobial activity. AUC measures the total drug exposure over time. Consistent AUC values across both routes and doses indicate the overall bioavailability and therapeutic exposure, essential for determining dosing regimens. The half-life reflects the time required for the plasma concentration to decrease by half. Longer half-lives in infusion suggest prolonged drug presence in the system, which may support less frequent dosing but requires careful monitoring for prolonged exposure effects. Tmax indicates the time taken to reach Cmax. Short Tmax in bolus administration highlights rapid absorption, while longer Tmax in infusion reflects gradual release and distribution, important for timing therapeutic effects. Ka (Absorption Rate Constant) of 0.0 indicates that oritavancin is administered directly into the bloodstream, bypassing the absorption phase typical of oral or subcutaneous routes. Vss indicates the extent of drug distribution in the body relative to the plasma. Higher values suggest extensive distribution into tissues, important for treating infections in various body compartments. Clearance measures the body’s efficiency in eliminating the drug. Similar clearance rates across doses and routes suggest consistent elimination mechanisms. AFE (Absolute Fractional Elimination) values of 0.0 indicate that the fraction of the drug eliminated per unit time remains constant across different doses and routes.

About 85% of oritavancin is bound to plasma proteins indicating that free (active) drug concentrations will be significantly less. The Vd of oritavancin is estimated at 87.6 L, suggesting extensive tissue distribution. Oritavancin has a prolonged terminal half-life and is excreted unchanged in both urine and feces [8, 54]. A pharmacokinetic study reported that Oritavancin displays multiexponential decline and has a terminal half-life a terminal half-life ranging from 135.8 to 273.8 h [55]. Dose-ranging studies have shown that this antibiotic displays linear pharmacokinetics with renal excretion being the major route of elimination and no dose adjustment is required on the basis of age or renal function or for patients with moderate hepatic impairment [56]. Following a 1200-mg dose (over 3 h) of oritavancin, the terminal half-life is predicted to start at concentrations of 5–20 µg/mL [57]. In a rabbit model of meningitis induced by Streptococcus pneumoniae, a single dose of oritavancin demonstrated a reduction in bacterial titers in cerebrospinal fluid (CSF) which was efficiently comparable to that of continuous infusion of ceftriaxone [58]. Oritavancin’s pharmacokinetic profile in rabbits demonstrated rapid distribution to bone tissues, where concentrations remained stable for up to 168 h [59]. Overall, the PBPK results of this study corroborate previous preclinical and clinical reports on oritavancin in term of half-life (147–158 h by infusion) and 60 days (1440 h) for total clearance [4, 8].

Limitations and strengths of this study

The limitations of this study are: (i) it relies heavily on computational approaches, which, while powerful, may not capture all the complexities of biological systems, thus experimental validation is necessary to confirm the predictions. (ii) Oritavancin’s moderate solubility in water might pose formulation challenges that were not addressed in this study. (iii) The study identified that Oritavancin is not permeable across the blood-brain barrier, potentially limiting its effectiveness against central nervous system infections. (iv) Although the PBPK model used AI-assisted data, the accuracy of predictions can still vary depending on the quality and extent of input data.

The strengths of this study are: (i) A combination of target prediction, molecular docking, molecular dynamics simulation, and PBPK modeling were used which provide a robust and multifaceted analysis. (ii) Identification of seven molecular targets in both bacteria and humans offers a broad understanding of Oritavancin’s potential interactions. (iii) Detailed molecular docking and dynamics simulation results revealed high binding affinities and stability with key proteins, suggesting potential mechanisms of action. (iv) The integration of physicochemical properties with physiological parameters in the PBPK model allows for accurate prediction of Oritavancin’s behavior in the body. (v) The PBPK model’s validation with clinical pharmacokinetic data enhances the reliability of the study’s predictions.

Future perspectives of this study

Future studies should include: (i) experimental validation of the computational predictions to confirm molecular targets and binding interactions. (ii) research focusing on improving the solubility and formulation of Oritavancin to enhance its clinical efficacy. (iii) investigating Oritavancin’s potential for treating a wider range of infections, including those involving the central nervous system, could expand its clinical utility, (iv) comparing Oritavancin with other antibiotics using similar computational approaches could provide insights into its relative advantages and potential synergies, and (v) further clinical trials to explore the therapeutic potential and safety profile of Oritavancin in diverse patient populations and infection types.

Conclusion

This study provides a comprehensive insilico analysis of Oritavancin, encompassing various computational approaches. The integration of these computational approaches has yielded several significant insights. The target prediction identified multiple potential protein targets for oritavancin. These targets include crucial bacterial cell wall synthesis enzymes and other proteins essential for bacterial cell wall and proliferation. The docking studies revealed strong binding affinities of oritavancin to its predicted targets. The binding modes provided insights into the mechanism of action, confirming the drug’s high specificity and potency against gram-positive bacteria. By leveraging the capabilities of ChatGPT, the study successfully simulated the pharmacokinetic profiles of Oritavancin. Together, these findings underscore the efficacy of oritavancin and its potential to combat resistant bacterial strains. Overall, this study has showcased oritavancin’s pharmacological profile and provided a robust framework for its clinical application. Basically, this study has ascertained the suitability of 200 mg of oritavancin per day for 3 days by infusion or injection to be safe scheme. The methodologies and insights presented herein pave the way for the development of more effective antibacterial therapies and highlight the potential of computational approaches in drug discovery and development.