Synergistic Potential of Repurposing non-β-lactam Compounds as Class A serine β-lactamases Inhibitor: Insights from MolecularDocking, Molecular Dynamics Simulations and Antimicrobial Potentiation

Abstract

The global dissemination of extended-spectrum β-lactamases (ESBLs) represents an urgent public health challenge. This study employed a drug repurposing strategy integrating in-silico screening, molecular dynamics (MD) simulations, and in-vitro validation to identify FDA-approved compounds capable of potentiating β-lactam antibiotics against ESBL-producing bacteria. Structure-based virtual screening of 400 FDA-approved compounds against SHV-1 β-lactamase (PDB: 4ZAM) identified 28 compounds with promising binding energies. Four commercially available compounds; epinephrine, omeprazole, sulfadimethoxine, and captopril exhibited binding energies ranging from − 8.15 to − 9.58 kcal/mol with RMSD values of 1.5–2.3 Å, comparable to the reference inhibitor avibactam. MD simulations (250 ns) confirmed the stability of all protein–ligand complexes, with the SHV-1/epinephrine complex demonstrating the lowest RMSD (0.155 nm) and most compact structure. In-vitro evaluation against an ESBL-producing Escherichia coli clinical isolate revealed that epinephrine and omeprazole effectively enhanced antibiotic activity in disk diffusion assays. The ceftriaxone/omeprazole combination achieved the highest potentiation (20 mm at 250 µg/mL), followed by ceftriaxone/epinephrine (18 mm at 250 µg/mL) and cefuroxime/omeprazole (19 mm at 125 µg/mL). Notably, captopril showed no in-vitro activity despite favorable computational predictions, underscoring the importance of experimental validation in drug discovery. These findings highlight the potential of epinephrine and omeprazole as readily available adjuvants against Class A serine β-lactamases, offering a cost-effective strategy to combat ESBL-mediated antibiotic resistance.

Introduction

Antibiotics have long been regarded as the cornerstone of modern medicine in combating diseases caused by microbial infections. In recent decades, however, the widespread misuse and overuse of antibiotics have accelerated the emergence of antimicrobial resistance (AMR) worldwide, posing a critical global public health threat¹. According to recent estimates, approximately 700,000 people die annually from infections caused by AMR pathogens, a figure projected to escalate dramatically to 10 million deaths per year by 2050 if current trends persist². This alarming trajectory underscores the urgent need for innovative strategies to combat resistant bacterial infections.

Among all antimicrobial agents, the β-lactam antibiotic family remains the primary therapeutic choice for treating pathogenic bacterial infections, particularly in critical care settings, accounting for approximately 65% of total antibiotic consumption due to its excellent safety profile and broad-spectrum efficacy^3,4. Structurally, β-lactam antibiotics are classified into four distinct groups: penicillins, cephalosporins, carbapenems, and monobactams⁵. Mechanistically, these drugs exert their bactericidal activity by binding to penicillin-binding proteins (PBPs), which are essential enzymes for bacterial cell wall biosynthesis. This interaction inhibits PBP transpeptidase activity, thereby disrupting peptidoglycan cross-linking and ultimately leading to cell lysis and death⁴.

Bacteria have evolved multiple resistance mechanisms against β-lactam antibiotics, including modification of PBP target sites, enhanced expression of active efflux pumps, reduced outer membrane permeability, and enzymatic degradation of the antibiotic. Among these, the production of β-lactamase enzymes represents the most prevalent and clinically significant resistance mechanism. These enzymes hydrolyze the β-lactam ring through an acylation/deacylation-based catalytic process, thereby rendering the antibiotic inactive⁶. Currently, diverse classes of β-lactamases have been identified in clinical isolates, including penicillinases, extended-spectrum β-lactamases (ESBLs), AmpC cephalosporinases, metallo-β-lactamases (MBLs), and Klebsiella pneumoniae carbapenemases (KPCs)^6,7.

Among the various β-lactamase classes, ESBL-producing bacteria have emerged as pathogens of critical clinical importance and have garnered significant attention from the scientific community^8,9. According to the Ambler molecular classification system, ESBLs belong to Class A serine β-lactamases and are characterized by their ability to hydrolyze a broad range of β-lactam antibiotics, including penicillins, aztreonam, and first-, second-, and third-generation cephalosporins^6,10. The most clinically prevalent ESBL subtypes are the plasmid-encoded TEM and SHV β-lactamases, which can be inhibited by classical β-lactamase inhibitors such as clavulanic acid, tazobactam, and sulbactam^9,11. The SHV-1 β-lactamase, in particular, serves as the archetypal Class A enzyme and represents an ideal model for investigating novel inhibitor compounds due to its well-characterized three-dimensional structure and catalytic mechanism^12,13. The crystal structure of SHV-1 in complex with avibactam (PDB identifier: 4ZAM) provides valuable structural insights into the active site architecture and has facilitated numerous structure-based drug design studies targeting this enzyme class¹⁴.

The rapid global dissemination of ESBL-producing bacteria in clinical settings has escalated into a major public health crisis⁸. In 2017, the World Health Organization (WHO) published a priority pathogens list identifying 12 bacterial families that pose the greatest threat to human health due to their antibiotic resistance profiles. This list prominently features Gram-negative pathogens, including Acinetobacter baumannii, Pseudomonas aeruginosa, Klebsiella pneumoniae, and Escherichia coli, many of which are prolific ESBL producers¹⁵. Notably, ESBL-producing Enterobacteriaceae have been classified among the “critical” priority group, emphasizing the urgent need for new therapeutic interventions.

Given the declining pipeline of novel antibiotics and the substantial time and financial investment required for de novo drug development, alternative strategies to preserve and extend the efficacy of existing β-lactam antibiotics have gained considerable attention¹⁶. One particularly promising approach involves the development of novel β-lactamase inhibitors that can be co-administered with β-lactam antibiotics to protect them from enzymatic degradation¹⁷. While traditional β-lactamase inhibitors such as clavulanic acid contain a β-lactam core structure and are themselves susceptible to certain resistance mechanisms, non-β-lactam inhibitors offer potential advantages including resistance to hydrolysis by metallo-β-lactamases and novel mechanisms of action¹⁸. The recent approval of Avibactam, a diazabicyclooctane-based non-β-lactam inhibitor, has validated this therapeutic strategy and demonstrated its clinical utility against multidrug-resistant infections^19,20.

Drug repurposing, also known as drug repositioning, represents an attractive strategy for rapidly identifying new therapeutic applications for existing approved medications. This approach offers several advantages over traditional drug discovery, including reduced development timelines, lower costs, established safety profiles, and existing manufacturing processes²¹. By screening libraries of FDA-approved compounds against novel therapeutic targets, researchers can potentially identify adjuvant molecules capable of restoring antibiotic susceptibility in resistant bacterial strains²².

Several strategies have been proposed to improve β-lactam antibiotic stewardship, including the development of new antibiotics such as Cefiderocol, which received United States Food and Drug Administration (FDA) approval in 2019 for the treatment of complicated urinary tract infections caused by multidrug-resistant Gram-negative bacteria²³, and the identification of novel adjuvant compounds that can extend the therapeutic lifespan of existing antibiotics²⁴. Combination therapy using β-lactam antibiotics with adjuvant compounds has shown promise as a strategy to overcome resistance mechanisms and enhance antimicrobial efficacy²⁵.

The current study aims to evaluate the potential inhibitory activity of repurposed non-β-lactam drugs against Class A serine β-lactamases, with particular focus on SHV-1 β-lactamase, using an integrated computational and experimental approach. We employed structure-based virtual screening to identify candidate compounds from a library of FDA-approved drugs, followed by in-vitro validation of antimicrobial potentiation activity. For the initial proof-of-concept validation, we utilized a well-characterized ESBL-producing Escherichia coli clinical isolate, as E. coli represents the most frequently encountered ESBL-producing organism in clinical settings and serves as a reliable model for evaluating β-lactamase inhibitor activity²⁶. This single-strain approach was deliberately selected to establish initial efficacy before proceeding to broader screening across multiple bacterial species and clinical isolates, which will be addressed in future investigations. By combining in-silico molecular docking with in-vitro antimicrobial assays, this study seeks to identify readily available compounds capable of restoring β-lactam antibiotic efficacy against ESBL-producing pathogens.

Methods

In-silico analysis

Protein selection and preparation

The target β-lactamase protein was selected based on the Bush–Jacoby functional classification system, and its three-dimensional structure was retrieved from the RCSB Protein Data Bank (PDB; https://www.rcsb.org). This study focused on the Class A serine β-lactamase SHV-1, which represents one of the most clinically prevalent ESBL enzymes. The coordinates and structure factors for the SHV-1/avibactam complex were obtained from the PDB (identifier: 4ZAM) at a resolution of 1.42 Ź⁴. The crystal structure was prepared using the Structure Preparation module in Molecular Operating Environment (MOE) software version 2024.0601²⁷. During protein preparation, missing hydrogen atoms were added to their standard geometry, and protonation states were assigned using the Protonate 3D tool at physiological pH 7.4. All crystallographic water molecules were removed to focus on direct protein–ligand interactions. The structure was subsequently energy-minimized using the MMFF94x force field, with tethering constraints applied to heavy atoms (force constant: 100 kcal mol⁻¹ Å⁻²) to maintain the experimental conformation while optimizing hydrogen atom positions. The final prepared structure was saved for subsequent docking calculations.

Ligand library selection and preparation

A structure-based virtual screening approach was employed to identify potential non-β-lactam inhibitors of SHV-1. The reference compound avibactam, a clinically approved diazabicyclooctane β-lactamase inhibitor with established activity against Class A enzymes, was used as the positive control. The avibactam structure was retrieved from the PubChem database (CID: 9835049) and served as the query molecule for similarity-based compound selection²⁸. A library of 400 FDA-approved compounds was obtained from the SwissSimilarity database (http://www.swisssimilarity.ch) by entering the avibactam SMILES notation and selecting the “FDA-combined drug” option²⁹. This approach was chosen to identify structurally related compounds with established safety profiles suitable for drug repurposing. All retrieved compounds were imported into MOE, subjected to 3D protonation at pH 7.4, and energy-minimized using the MMFF94x force field to a root-mean-square (RMS) gradient of 0.1 kcal mol⁻¹ Å⁻¹. The prepared ligand library was saved as an MDB database file for molecular docking calculations.

Molecular docking protocol

Molecular docking calculations were performed using the molecular operating environment (MOE) software version 2024.0601²⁷. The prepared ligand library comprising 400 FDA-approved compounds structurally related to avibactam, was used for molecular docking analysis. The co-crystallized ligand avibactam was included as the reference standard for docking validation and binding mode comparison. All ligands were compiled into a single molecular database (MDB) file for batch docking calculations. The binding site of SHV-1 β-lactamase was defined using the MOE Site Finder module, centered on the co-crystallized avibactam position within the active site cavity. All crystallographic water molecules and non-essential cofactors were removed prior to docking. The protein structure was prepared with hydrogen atoms added and partial charges assigned using the MMFF94x force field parameters. Docking simulations were executed using the MOE Dock module with the following parameters: the Triangle Matcher algorithm was employed for initial ligand placement to generate diverse binding poses within the active site. Pose refinement was performed using the Forcefield method with rigid receptor treatment. A two-stage scoring approach was applied: London dG was used for initial pose ranking during placement, followed by GBVI/WSA dG rescoring for final binding affinity estimation, which accounts for solvation effects and provides more accurate binding free energy predictions^30,31. For each compound, 100 docking poses were generated, from which the top 30 poses were retained based on binding scores. The docking protocol was validated by re-docking the co-crystallized avibactam ligand into the SHV-1 active site. The root-mean-square deviation (RMSD) between the docked pose and the crystallographic position was calculated to assess docking accuracy; an RMSD value ≤ 2.0 Å was considered acceptable. Following validation, the top 10 poses for each test compound were selected for detailed interaction analysis. Binding interactions, including hydrogen bonds, van der Waals contacts, and π-interactions with active site residues, were analyzed and visualized using the MOE Ligand Interactions module. Compounds were ranked based on binding scores, RMSD values, and the quality of interactions with key catalytic residues (Ser70, Ser130, Lys73, and Arg244).

Molecular dynamics simulation analysis

Molecular dynamics (MD) simulations were performed to evaluate the dynamic stability and binding behavior of the top-ranked docked complexes of the selected compounds with SHV-1 β-lactamase using the GROMACS 2025 software suite³². The SHV-1/avibactam complex was also simulated as a reference control for comparative analysis, given that avibactam is a clinically approved β-lactamase inhibitor with well-characterized binding properties. System preparation and topology generation were carried out using the CHARMM-GUI web server (https://www.charmm-gui.org)³³. The CHARMM36m force field was applied for protein parameterization, while the CGenFF (CHARMM General Force Field) was employed for ligand topology generation with appropriate bond orders, partial charges, and geometric parameters. Each protein–ligand complex was solvated in a cubic simulation box with TIP3P water molecules, maintaining a minimum distance of 1.0 nm from the protein surface to the box edge, and periodic boundary conditions were applied in all three dimensions. The systems were neutralized and brought to physiological ionic strength (0.15 M) by adding appropriate numbers of Na⁺ and Cl⁻ counter ions. Energy minimization was performed using the steepest descent algorithm for a maximum of 50,000 steps with a force convergence criterion of 1000 kJ mol⁻¹ nm⁻¹ to remove steric clashes and relax the initial structures. Equilibration was conducted in two sequential phases. First, the systems were equilibrated under the NVT (constant number of particles, volume, and temperature) ensemble for 1 ns to stabilize the temperature at 300 K using the V-rescale thermostat with a coupling constant of 0.1 ps. Subsequently, NPT (constant number of particles, pressure, and temperature) ensemble equilibration was performed for an additional 1 ns to equilibrate pressure at 1.0 bar using the Parrinello-Rahman barostat with a coupling constant of 2.0 ps. During both equilibration phases, position restraints (force constant: 1000 kJ mol⁻¹ nm⁻²) were applied to the protein and ligand heavy atoms. Production MD simulations were performed for 250 ns under NPT conditions without positional restraints, using a 2 fs integration time step. Bond lengths involving hydrogen atoms were constrained using the LINCS (Linear Constraint Solver) algorithm. Short-range non-bonded interactions were calculated with a cut-off distance of 1.2 nm, while long-range electrostatic interactions were computed using the Particle Mesh Ewald (PME) method with a Fourier grid spacing of 0.16 nm. Trajectory analysis was performed using GROMACS built-in tools. Key structural parameters, including root mean square deviation (RMSD), root mean square fluctuation (RMSF), radius of gyration (Rg), and solvent-accessible surface area (SASA), were calculated to evaluate conformational changes, residue-level flexibility, and structural compactness of the protein–ligand complexes during the simulation. Hydrogen bond interactions between SHV-1 and the ligands were examined using geometric criteria of donor-acceptor distance ≤ 0.35 nm and donor-hydrogen-acceptor angle ≤ 30°³⁴. Trajectory snapshots and analytical outputs were processed and visualized using XMGrace software for graphical plots and PyMOL for three-dimensional structural representations.

Pharmacokinetic considerations

Detailed in-silico ADMET predictions were not performed for the four selected compounds, as epinephrine, omeprazole, sulfadimethoxine, and captopril are FDA-approved drugs with well-established pharmacokinetic and safety profiles documented in regulatory databases and published literature. The known clinical safety of these compounds supports their suitability for drug repurposing applications. Epinephrine is approved for emergency treatment of anaphylaxis and cardiac arrest³⁵; Omeprazole is a widely prescribed proton pump inhibitor³⁶; Sulfadimethoxine is an approved veterinary sulfonamide antibiotic³⁷; and Captopril is a first-generation ACE inhibitor used for hypertension management³⁸. The extensive clinical experience with these drugs eliminates the need for predictive ADMET modeling typically required for novel chemical entities.

In-vitro analysis

Selection of bacterial strain

A clinical isolate of ESBL-producing Escherichia coli was obtained from the microbiology laboratory at Omdurman Military Hospital, Khartoum, Sudan. The isolate was identified using conventional microbiological methods, including Gram staining and standard biochemical tests. The IMViC test pattern (indole positive, methyl red positive, Voges-Proskauer negative, citrate negative) confirmed the identity as E. coli³⁹. The isolate was screened for ESBL production using the double-disc synergy test (DDST) with ceftriaxone (30 µg) and ceftriaxone/clavulanic acid (30/10 µg) discs, following the Clinical and Laboratory Standards Institute (CLSI) M100 guidelines⁴⁰. A zone diameter difference of ≥ 5 mm for the ceftriaxone/clavulanic acid disc compared to ceftriaxone alone was interpreted as ESBL-positive. The confirmed ESBL-producing strain demonstrated resistance to cefuroxime (2 mm zone) and ceftriaxone (6 mm zone) while remaining susceptible to meropenem (20 mm zone), consistent with the typical phenotypic profile of ESBL-producing Enterobacteriaceae (Table 1). Quality control was performed using reference strains: Escherichia coli ATCC 25,922 (ESBL-negative control) and Klebsiella pneumoniae ATCC 700,603 (ESBL-positive control) to validate the ESBL detection method and antimicrobial susceptibility testing procedures⁴¹.

Table 1.

Molecular docking result of selected compounds against SHV-1.

Compounds name	S (kcal/mol)	RMSD	Hydrogen bond (distance)
Avibactam	− 10.7	2	Arg244(2.49), Asn170(2.90), Asn132(2.47), Ser130(2.97), Ala237 (2.95, 2.65), Ser70 (2.13)
Captopril	− 9.58	2.3	Asn170(2.62), Asn132(2.01, 2.21), Ala237(2.38)
Epinephrine	− 8.97	1.5	Ala237(2.18, 2.28), Asn132(2.93), Asp104(2.53)
Omeprazole	− 8.25	1.8	Asp104(2.13), Asn132(2.09)
Sulfadimethoxine	− 8.15	2.1	Asn132(2.42), Thr235(2.46)

*RMSD root-mean-square deviation.

Selection of chemical compounds

From the virtual screening campaign, 28 compounds demonstrated promising binding energies (≤ − 8.0 kcal/mol) and acceptable RMSD values (≤ 2.5 Å) comparable to the reference inhibitor avibactam. However, due to limited commercial availability and accessibility constraints, only four of these top-ranked hits could be procured for in-vitro validation studies. The selected compounds: epinephrine, omeprazole, sulfadimethoxine, and captopril were obtained in pharmaceutical-grade form (100% purity) from Tabuk Pharmaceutical Co. Ltd. (Sudan). Stock solutions were prepared by dissolving each compound in dimethyl sulfoxide (DMSO) and stored at 4 °C until use. Working solutions were prepared immediately before each experiment using an analytical balance and serial dilution.

In-vitro testing of the selected compounds

The intrinsic antibacterial activity of the selected compounds was evaluated using the disc diffusion method and agar well diffusion (cup-plate) technique to determine whether the test compounds possess direct antimicrobial properties independent of β-lactamase inhibition. Mueller-Hinton Agar (MHA; Oxoid Ltd., Hampshire, England) was used as the culture medium for all antimicrobial susceptibility testing.

The bacterial inoculum was prepared by suspending fresh overnight colonies of the ESBL-producing Escherichia coli isolate in sterile physiological saline (0.85% NaCl) and adjusting the turbidity to 0.5 McFarland standard (approximately 1.5 × 10⁸ CFU/mL) using a densitometer. The McFarland standard was prepared using 1% BaCl₂ and 1% H₂SO₄ solutions. Stock solutions of each test compound were prepared by dissolving 1 mg in 1 mL of DMSO to yield a concentration of 1000 µg/mL, and working solutions were subsequently prepared by serial dilution to obtain three test concentrations: 1000, 500, 250 and 125 µg/mL. A uniform bacterial lawn was prepared by spreading the standardized inoculum onto MHA plates using sterile cotton swabs. For disc diffusion testing, sterile filter paper discs (6 mm diameter; Whatman International Ltd., England) were impregnated with 125 µL, 250 µL, 500 µL, or 1000 µL of each test compound at the specified concentrations. Four discs were placed per plate with center-to-center spacing of 24 mm to prevent zone overlapping.

Controls

DMSO alone (without test compounds) served as the negative control to assess potential solvent effects on bacterial growth. Standard antibiotic discs (ceftriaxone 30 µg, cefuroxime 30 µg, and meropenem 10 µg) were included as reference controls for susceptibility testing.

The plates were incubated aerobically at 37 °C for 24 h. Zones of inhibition were measured in millimeters using a calibrated ruler. All antimicrobial assays were performed in quadruplicate to quintuplicate (4–5 independent experiments), and results are expressed as mean inhibition zone diameter ± standard deviation (SD).

In-vitro testing of selected compounds in combination with cefuroxime, ceftriaxone, and anti-microbial activity determination

The potential of the selected compounds to restore β-lactam antibiotic efficacy was evaluated by testing them in combination with cefuroxime and ceftriaxone against the ESBL-producing Escherichia coli isolate. The anti-microbial activity was determined using the agar well diffusion (cup-plate) method and confirmed by the modified E-test technique. The bacterial inoculum was prepared in Tryptic Soy Broth (TSB) following CLSI M100 guidelines, with the optical density at 600 nm (OD₆₀₀) adjusted to 0.08–0.12, corresponding to approximately 1.5 × 10⁸ CFU/mL. The test compounds were serially diluted to obtain the following concentrations: 2000, 1000, 500, 250, 125, 62.5, 31.25, 15.625, and 7.825 µg/mL. Each compound at varying concentrations was tested in combination with cefuroxime (30 µg) and ceftriaxone (30 µg) to evaluate their inhibitory effect against the ESBL-producing isolate. The plates were incubated at 37 °C for 18–24 h, and zones of inhibition were measured in millimeters to determine the anti-microbial activity. For the agar well diffusion (cup-plate) method, wells of 6 mm diameter were prepared using a sterile cork borer, and 50–100 µL of the test solution was dispensed into each well. The test compounds were added to wells adjacent to the antibiotic discs to evaluate combination effects. Controls for combination testing included: (1) antibiotic alone (to confirm resistance phenotype), (2) compound alone (to confirm absence of direct antimicrobial activity), and (3) DMSO control (to assess solvent effects). Susceptibility interpretation was performed according to CLSI M100 breakpoints for Enterobacteriaceae: For ceftriaxone, susceptible (S) ≥ 23 mm, intermediate (I) 20–22 mm, resistant (R) ≤ 19 mm; for cefuroxime, S ≥ 18 mm, I 15–17 mm, R ≤ 14 mm. Although broth microdilution is the CLSI gold-standard method for MIC determination, the agar well diffusion method was employed in this proof-of-concept study as a practical screening approach for evaluating antimicrobial potentiation effects.

Results and Discussion

Molecular docking

In molecular docking analysis, more negative binding energy values indicate stronger predicted binding affinity due to greater thermodynamic stability of the ligand–protein complex⁴². The reliability of the molecular docking protocol was first assessed by re-docking the co-crystallized ligand avibactam into the active site of SHV-1 β-lactamase (PDB: 4ZAM). The re-docked pose successfully reproduced the crystallographic binding conformation with an RMSD of 2.0 Å, which falls within the acceptable threshold (≤ 2.5 Å) for validating docking accuracy. Avibactam, a clinically approved diazabicyclooctane-based β-lactamase inhibitor⁴³, exhibited a binding score of − 10.7 kcal/mol and established an extensive hydrogen bonding network with six key active site residues: Ser70 (2.13 Å), Ser130 (2.97 Å), Asn132 (2.47 Å), Asn170 (2.90 Å), Ala237 (2.95 and 2.65 Å), and Arg244 (2.49 Å) (Fig. 1A). The interaction with Ser70, the catalytic nucleophile, is particularly significant as avibactam forms a covalent acyl-enzyme intermediate with this residue, resulting in reversible β-lactamase inactivation. The interactions with Ser130, part of the conserved SDN motif (Ser130-Asp131-Asn132), may influence proton transfer during catalysis. These validated docking parameters were subsequently employed for virtual screening of the compound library.

Fig. 1.

Docked poses of leading compounds with SHV1. (A–C) Avibactam: (A) overall 3D view of the SHV1 binding site, (B) zoomed-in view highlighting key interacting residues, and (C) 2D interaction diagram. (D–F) Captopril: (D) overall 3D binding site view, (E) zoomed-in view of the binding region, and (F) 2D representation of protein–ligand interactions. (G–I) Epinephrine: (G) 3D view of the RUNX1 binding pocket, (H) zoomed-in view emphasizing interacting residues, and (I) 2D interaction diagram. (J-L) Omeprazole: (J) overall 3D view of the binding site, (K) zoomed-in view of key contacts within the pocket, and (L) 2D interaction representation. (M–O) Sulfadimethoxine: (M) overall 3D view of the binding site, (N) zoomed-in view of key contacts within the pocket, and (O) 2D interaction representation of Sulfadimethoxine bound to SHV1.

Structure-based virtual screening of 400 FDA-approved compounds against the SHV-1 active site identified 28 compounds exhibiting promising binding energies (≤ − 8.0 kcal/mol) and acceptable RMSD values (≤ 2.5 Å) comparable to the reference inhibitor avibactam. From these 28 hits, four compounds representing distinct chemical and pharmacological classes were prioritized for in-vitro validation based on commercial availability and structural diversity. Epinephrine (adrenaline) is an endogenous catecholamine and a non-selective adrenergic receptor agonist widely used in emergency medicine for the treatment of anaphylaxis, cardiac arrest, and severe asthma, with its catechol ring system containing hydroxyl groups capable of hydrogen bond donation that may facilitate interactions with polar residues in the β-lactamase active site³⁵. Omeprazole is a benzimidazole-based proton pump inhibitor (PPI) that irreversibly inhibits the gastric H⁺/K⁺-ATPase and is extensively prescribed for the management of gastroesophageal reflux disease, peptic ulcers, and Helicobacter pylori eradication therapy, with its sulfinyl group and benzimidazole moiety providing multiple sites for potential hydrogen bonding and hydrophobic interactions³⁶. Sulfadimethoxine is a long-acting sulfonamide antibiotic that inhibits bacterial dihydropteroate synthase, thereby disrupting folate biosynthesis, and although it possesses intrinsic antibacterial activity, its potential role as a β-lactamase inhibitor has not been previously explored, with its sulfonamide functional group and dimethoxypyrimidine ring offering diverse interaction possibilities within the enzyme active site. Captopril is a thiol-containing angiotensin-converting enzyme (ACE) inhibitor used in the treatment of hypertension and heart failure, with the presence of a free thiol group that is responsible for zinc coordination in ACE inhibition potentially enabling unique interactions with the SHV-1 active site that differ from the other test compounds³⁸. The selection of these four structurally diverse compounds enabled evaluation of different chemical scaffolds as potential β-lactamase inhibitor candidates, providing preliminary structure-activity relationship insights.

The molecular interactions between the selected compounds and SHV-1 β-lactamase were analyzed in detail to understand their binding modes and compare them with the reference inhibitor avibactam (Table 2; Fig. 1). Captopril demonstrated the most favorable binding affinity among the test compounds, with a binding score of − 9.58 kcal/mol and an RMSD of 2.3 Å. The compound established three hydrogen bonds with active site residues: a single hydrogen bond with Asn170 (2.62 Å), two hydrogen bonds with Asn132 (2.01 and 2.21 Å), and one hydrogen bond with Ala237 (2.38 Å). The short hydrogen bond distances with Asn132 (2.01 and 2.21 Å) indicate strong interactions, as distances below 2.5 Å are typically associated with high-affinity binding. The thiol group of captopril oriented toward the hydrophobic region of the active site, while the proline carboxylate moiety extended toward the solvent-accessible surface. Notably, captopril shared three key interaction residues (Asn132, Asn170, and Ala237) with avibactam, suggesting a partially overlapping binding mode that may contribute to competitive inhibition of the enzyme. Epinephrine exhibited a binding score of − 8.97 kcal/mol with the lowest RMSD value (1.5 Å) among all test compounds, indicating a highly stable and well-defined binding pose within the active site. The compound formed four hydrogen bonds involving Ala237 (2.18 and 2.28 Å), Asn132 (2.93 Å), and Asp104 (2.53 Å). The catechol hydroxyl groups of epinephrine served as hydrogen bond donors, with the meta-hydroxyl group interacting with Ala237 and the para-hydroxyl group forming contacts with Asn132. The secondary amine group established additional polar contacts with Asp104. The dual hydrogen bonds with Ala237 (2.18 and 2.28 Å) are particularly noteworthy, as this residue is located adjacent to the catalytic Ser70 and contributes to substrate positioning during β-lactam hydrolysis. The compact molecular structure of epinephrine allowed optimal accommodation within the active site cavity, as reflected by the low RMSD value. Omeprazole displayed a binding score of − 8.25 kcal/mol with an RMSD of 1.8 Å. Despite forming only two hydrogen bonds, the interactions exhibited short bond distances indicative of strong binding: Asp104 (2.13 Å) and Asn132 (2.09 Å). The benzimidazole ring system of omeprazole occupied a hydrophobic pocket within the active site, establishing van der Waals contacts with surrounding non-polar residues. The sulfinyl group oriented toward the polar region, enabling hydrogen bond formation with Asp104. The pyridine ring extended toward the entrance of the active site channel, potentially contributing to steric blockade of substrate access. Although omeprazole formed fewer hydrogen bonds compared to other compounds, the short bond distances (≤ 2.13 Å) and favorable positioning of the benzimidazole scaffold suggest effective complementarity with the SHV-1 binding site. Sulfadimethoxine exhibited a binding score of − 8.15 kcal/mol with an RMSD of 2.1 Å. The compound formed two hydrogen bonds with Asn132 (2.42 Å) and Thr235 (2.46 Å). The sulfonamide group of sulfadimethoxine positioned within the central region of the active site pocket, with the nitrogen atom serving as a hydrogen bond donor to Asn132. The dimethoxypyrimidine ring extended toward the peripheral region of the binding site, where it established a hydrogen bond with Thr235. This interaction with Thr235, located adjacent to the conserved Ala237, may interfere with the positioning of the β-lactam substrate during catalysis. The aniline moiety occupied a shallow hydrophobic groove, contributing additional van der Waals stabilization to the complex. Comparative analysis of the binding interactions revealed several important findings regarding the potential inhibitory mechanisms of the selected compounds. All four test compounds shared hydrogen bonding with Asn132, a highly conserved residue in Class A β-lactamases that plays a critical role in substrate recognition, transition state stabilization, and proton transfer during the catalytic cycle, suggesting that Asn132 represents a key anchor point for non-β-lactam inhibitor binding in the SHV-1 active site. Additionally, two compounds (epinephrine and omeprazole) formed hydrogen bonds with Asp104, a residue located at the entrance of the active site channel that may influence substrate access and binding kinetics. Importantly, unlike avibactam, none of the four test compounds formed direct hydrogen bonds with Ser70, the catalytic nucleophile responsible for the initial nucleophilic attack on the β-lactam carbonyl carbon, suggesting that these compounds may function as competitive, non-covalent inhibitors that occlude the active site and prevent substrate binding, rather than covalent inactivators that form stable acyl-enzyme intermediates. This mechanistic distinction has important implications for inhibitor potency and reversibility, as non-covalent inhibitors typically exhibit faster dissociation rates compared to covalent inhibitors. The binding scores of the four selected compounds ranged from − 8.15 to − 9.58 kcal/mol, which were 1.1 to 2.5 kcal/mol less favorable than avibactam (− 10.7 kcal/mol); however, these values fall within the range typically associated with moderate to strong binding affinity and are comparable to other reported non-β-lactam β-lactamase inhibitors in the literature. The consistently low RMSD values (1.5–2.3 Å) across all compounds indicated stable and reproducible docking poses, supporting the reliability of the predicted binding modes. Collectively, these in-silico findings provided a strong rationale for advancing the four selected compounds to in-vitro antimicrobial potentiation studies.

Table 2.

Antimicrobial susceptibility profile of the ESBL-producing Escherichia coli isolate.

Escherichia. coli	Ceftriaxone (30 µg/mL)	Cefuroxime (30 µg/mL)	Meropenem (10 µg/mL)
	6 mm (R)	2 mm (R)	20 mm (S)

*R resistance *S sensitive.

Molecular dynamics simulation analysis

To evaluate the conformational stability of SHV-1 β-lactamase and protein–ligand complexes, we performed molecular dynamics (MD) simulations for SHV-1/avibactam (reference), SHV-1/captopril, SHV-1/epinephrine, SHV-1/omeprazole, and SHV-1/sulfadimethoxine complexes. Our objective was to investigate their dynamic behavior by assessing key analysis parameters, including root mean square deviation (RMSD), root mean square fluctuation (RMSF), radius of gyration (Rg), solvent-accessible surface area (SASA), and hydrogen bond interactions. Through the analysis of these parameters, we aimed to gain valuable insights into the conformational dynamics and overall stability of the protein–ligand complexes. After completing the 250 ns simulations for all complexes, we conducted a comprehensive analysis of their trajectories. Specifically, we calculated the average RMSD values of the backbones for each protein–ligand complex. RMSD assesses protein stability by measuring deviations during MD simulation, where lower deviation indicates higher stability^44,45. A lower RMSD for a ligand-protein complex indicates successful binding pocket accommodation. The obtained average RMSD values for SHV-1/avibactam, SHV-1/captopril, SHV-1/epinephrine, SHV-1/omeprazole, and SHV-1/sulfadimethoxine were 0.1956 nm, 0.1623 nm, 0.155 nm, 0.1973 nm, and 0.2241 nm, respectively (Fig. 2A). The RMSD analysis demonstrated that all protein–ligand complexes-maintained stability and consistency throughout the simulation period with non-significant fluctuations. Notably, the SHV-1/epinephrine complex exhibited the lowest RMSD value (0.155 nm), followed by SHV-1/captopril (0.1623 nm), indicating that these compounds stabilize the protein conformation more effectively than the reference inhibitor avibactam. Further analysis showed that the average backbone fluctuations of all protein–ligand complexes were below 2.5 Å or 0.25 nm, suggesting that ligand binding did not significantly alter the backbone structure of the protein.

Fig. 2.

Molecular dynamics (MD) simulation analysis of SHV1 protein with leading compunds. (A) Root mean square deviation (RMSD) plots of SHV1 in complex with avibactam, captopril, epinephrine, omeprazole, and sulfadimethoxine showing the structural stability of each complex throughout the simulation. (B) Root mean square fluctuation (RMSF) plots of SHV1 complexes showing residue-wise flexibility and local conformational variations. (C) Graphical representation of radius of gyration (Rg) profiles of SHV1 in complex with avibactam, captopril, epinephrine, omeprazole, and sulfadimethoxine. (D) Solvent-accessible surface area (SASA) variation of SHV1 complexes throughout the simulation period. The analyses highlight the structural compactness and conformational stability of the protein–ligand complexes during molecular dynamics simulations.

To analyze the dynamic behavior of each amino acid residue within the protein–ligand complexes, we employed the root mean square fluctuation (RMSF) parameter^46,47. Figure 2B displays the RMSF plot for each complex, illustrating the variation of residues over time. The average RMSF values for SHV-1/avibactam, SHV-1/captopril, SHV-1/epinephrine, SHV-1/omeprazole, and SHV-1/sulfadimethoxine were 0.1223 nm, 0.1053 nm, 0.1093 nm, 0.1318 nm, and 0.1195 nm, respectively. After analysis of RMSF results, we found that all complexes exhibited stability with minimal fluctuations. The SHV-1/captopril complex demonstrated the lowest average RMSF (0.1053 nm), followed by SHV-1/epinephrine (0.1093 nm), indicating that these compounds effectively reduce the conformational flexibility of SHV-1, particularly in the active site region. Lower RMSF values signify better stability throughout the MD simulation. Therefore, the average RMSF values collectively suggest the stability of all protein–ligand complexes, implying favorable binding of the test compounds with SHV-1 β-lactamase.

The compactness of the protein–ligand complexes in a dynamic environment was assessed by analyzing the radius of gyration (Rg) parameter^48,49. Figure 2C depicts the relationship between Rg values of all complexes over the simulation time range. The trajectories of the complexes consistently demonstrated compactness, suggesting strong binding compatibility between each ligand and the active site of SHV-1. The average Rg values for SHV-1/avibactam, SHV-1/captopril, SHV-1/epinephrine, SHV-1/omeprazole, and SHV-1/sulfadimethoxine were 1.8319 nm, 1.822 nm, 1.8137 nm, 1.8277 nm, and 1.8348 nm, respectively. These values align with the acceptable range of approximately 1.8 ± 0.02 nm for stable globular proteins. The SHV-1/epinephrine complex exhibited the lowest Rg value (1.8137 nm), indicating the most compact protein structure. The analysis of Rg demonstrated that all protein–ligand complexes remained compact throughout the simulation, suggesting strong binding compatibility between the test compounds and the active site of SHV-1 β-lactamase.

To evaluate the energy linked to nonpolar solvation within the protein–ligand complexes during simulation, we explored the concept of solvent-accessible surface area (SASA). Nonpolar solvation models reliant on SASA can accurately anticipate protein structural conformations and protein–ligand binding affinities^49,50. In the present study, the SASA plot for the complexes demonstrates that the surface accessibility of the protein complexes remained stable throughout the simulation. The computed average SASA values for SHV-1/avibactam, SHV-1/captopril, SHV-1/epinephrine, SHV-1/omeprazole, and SHV-1/sulfadimethoxine were 122.55 nm², 122.69 nm², 119.39 nm², 128.2 nm², and 123.63 nm², respectively (Fig. 2D). The SHV-1/epinephrine complex exhibited the lowest SASA value (119.39 nm²), indicating reduced solvent exposure and a more buried binding interface. The obtained results indicate promising solvent accessibility for all complexes, implying minimal conformational alterations and stable binding between SHV-1 and the test compounds.

We calculated the average total number of hydrogen bonds formed throughout the MD simulation at various time points. For the complexes, the average number of hydrogen bonds was approximately 0.469 for SHV-1/avibactam, 0.3892 for SHV-1/captopril, 0.2716 for SHV-1/epinephrine, 0.7835 for SHV-1/omeprazole, and 0.3002 for SHV-1/sulfadimethoxine (Fig. 3). The SHV-1/omeprazole complex exhibited the highest average number of hydrogen bonds (0.7835), suggesting strong and persistent polar interactions throughout the simulation. This may be attributed to the presence of multiple polar functional groups in omeprazole capable of forming hydrogen bonds with active site residues. Interestingly, despite exhibiting fewer hydrogen bonds compared to avibactam, epinephrine demonstrated the lowest RMSD, Rg, and SASA values, suggesting that its binding stability may be primarily driven by optimal shape complementarity and hydrophobic interactions rather than extensive hydrogen bonding. These findings underscore the crucial role of hydrogen bond interactions in stabilizing the protein–ligand complexes while highlighting that different compounds employ diverse binding mechanisms to achieve stable association with SHV-1 β-lactamase.

Fig. 3.

Hydrogen bond interaction analysis of protein complex with selected compunds during molecular dynamics (MD) simulations. (A) Time-dependent hydrogen bond formation between SHV1 with avibactam, captopril, epinephrine, omeprazole, and sulfadimethoxine throughout the 2500 ns. The frequency and persistence of hydrogen bonds indicate the stability and strength of protein–ligand interactions during MD simulations.

In-vitro results of tested compounds

Epinephrine, Sulfadimethoxine, Omeprazole, and Captopril were assessed for intrinsic antibacterial activity against the ESBL-producing E. coli isolate. The compounds tested alone showed no antimicrobial effect (0 mm inhibition zones), except epinephrine and sulfadimethoxine, which exhibited weak antibacterial activity with inhibition zones of 9 mm and 7 mm, respectively, at a concentration of 1000 µg/mL. These results suggest that the compounds lack substantial inherent antibacterial properties at the tested concentrations. The DMSO negative control showed no inhibition of bacterial growth, confirming that observed effects were attributable to the test compounds.

Antimicrobial potentiation activity

The anti-microbial activity of the selected compounds in combination with β-lactam antibiotics was determined using the agar well diffusion method. Susceptibility categories were assigned according to CLSI M100 breakpoints for Enterobacteriaceae.

Ceftriaxone combinations: The highest antimicrobial potentiation was achieved by ceftriaxone/omeprazole combination (20 mm at 250 µg/mL), followed by ceftriaxone/epinephrine (18 mm at 250 µg/mL) and ceftriaxone/sulfadimethoxine (19 mm at 2000 µg/mL).

Cefuroxime combinations: The cefuroxime/omeprazole combination demonstrated the highest activity (19 mm at 125 µg/mL), followed by cefuroxime/epinephrine (18 mm at 125 µg/mL) and cefuroxime/sulfadimethoxine (14 mm at 250 µg/mL).

Notably, captopril showed no antimicrobial potentiation activity against the ESBL-producing isolate at any tested concentration, despite demonstrating favorable binding interactions in molecular docking studies (Table 3; Fig. 4).

Table 3.

Antibacterial activity of chemical compund/antibiotic combination against resistance isolation.

	Ceftriaxone (30 µg/mL)				Cefuroxime (30 µg/mL)
	compoundconc.				compoundconc.
Chemical compounds	2000 µg/mL	1000 µg/mL	500 µg/mL	250 µg/mL	1000 µg/mL	500 µg/mL	250 µg/mL	125 µg/mL	62.5 µg/mL	31.25 µg/mL	15.625 µg/mL	7.825 µg /mL
Epinephrine	23 mm (S)	20 mm (I)	19 mm (I)	18 mm (I)	20 mm (S)	18 mm (S)	18 mm (S)	18 mm (S)	16 mm (I)	16 mm (I)	13 mm (R)	13 mm (R)
Omeprazole	0 mm (R)	15 mm (I)	17 mm (I)	20 mm (I)	16 mm (I)	15 mm (I)	14 mm (I)	19 mm (S)	16 mm (I)	16 mm (I)	14 mm (I)	12 mm (R)
Sulfadimethoxine	19 mm (I)	10 mm (R)	8 mm (R)	6 mm (R)	14 mm (I)	11 mm (R)	14 mm (I)	10 mm (R)	10 mm (R)	10 mm (R)	10 mm (R)	5 mm (R)
Captopril	0 mm (R)	0 mm (R)	0 mm (R)	0 mm (R)	0 mm (R)	0 mm (R)	0 mm (R)	0 mm (R)	0 mm (R)	0 mm (R)	0 mm (R)	0 mm (R)

Significant values are in bold.

*R resistance *S sensitive.

Fig. 4.

Anti-microbial activity determination using the modified E-test technique. (A) Epinephrine in combination with cefuroxime; (B) Omeprazole in combination with cefuroxime; (C) Sulfadimethoxine in combination with cefuroxime against ESBL-producing E. coli.

Discussion

The emergence and global dissemination of ESBL-producing bacteria represent one of the most pressing challenges in contemporary antimicrobial therapy. The present study employed an integrated computational and experimental approach to identify FDA-approved drugs with potential β-lactamase inhibitory activity through drug repurposing. Our findings demonstrate that epinephrine and omeprazole, two structurally distinct compounds with well-established clinical safety profiles, can restore the antimicrobial efficacy of cephalosporins against ESBL-producing Escherichia coli.

The molecular docking analysis identified four compounds: Captopril, Epinephrine, Omeprazole, and Sulfadimethoxine with favorable binding energies ranging from − 8.15 to − 9.58 kcal/mol against SHV-1 β-lactamase, all within 1.1–2.5 kcal/mol of avibactam (− 10.7 kcal/mol), a clinically approved β-lactamase inhibitor. A key finding was the identification of Asn132 as a common interaction residue across all four test compounds. Asn132 is a highly conserved residue in Class A β-lactamases that plays essential roles in substrate recognition and catalytic function, suggesting it may serve as a critical anchor point for non-β-lactam inhibitor design⁵¹.

The molecular dynamics simulations provided crucial insights into the dynamic stability of the protein–ligand complexes. All five complexes maintained structural stability throughout the 250 ns simulation period, with average RMSD values below 0.25 nm. Remarkably, the SHV-1/epinephrine complex exhibited the lowest RMSD (0.155 nm), radius of gyration (1.8137 nm), and SASA values (119.39 nm²) among all complexes, including the reference avibactam complex, indicating superior binding stability. The SHV-1/omeprazole complex exhibited the highest average number of hydrogen bonds (0.7835), suggesting strong and persistent polar interactions. These computational observations correlated well with the subsequent in-vitro findings, where epinephrine and omeprazole demonstrated significant antimicrobial potentiation activity.

An intriguing discordance was observed between computational predictions and experimental outcomes for captopril. Despite exhibiting the highest binding affinity (− 9.58 kcal/mol) and favorable MD simulation parameters among the test compounds, Captopril showed no antimicrobial potentiation activity in-vitro at any tested concentration. Several factors may explain this discrepancy: captopril may bind to SHV-1 without effectively blocking the catalytic mechanism; the compound may exhibit poor membrane permeability limiting access to periplasmic β-lactamases; or the binding kinetics may be unfavorable for sustained enzyme inhibition under physiological conditions. This observation reinforces the critical importance of integrating computational predictions with experimental validation in drug discovery programs, as favorable binding affinity does not necessarily translate to functional enzyme inhibition.

In contrast, epinephrine and omeprazole demonstrated robust antimicrobial potentiation activity when combined with ceftriaxone and cefuroxime. The ceftriaxone/omeprazole combination achieved the highest activity (20 mm zone at 250 µg/mL), restoring susceptibility to a level comparable to the CLSI intermediate category. Although a classical β-lactamase inhibitor such as clavulanic acid was not included as a positive control in the present study, the primary objective was to assess comparative anti-microbial activity shifts relative to antibiotic alone under standardized conditions. Standard microbiological controls were incorporated to ensure experimental validity. The differential activity observed among the compounds likely reflects distinct mechanisms of action. Omeprazole, a benzimidazole derivative containing a sulfinyl functional group, may undergo nucleophilic interactions with active site residues, and the benzimidazole scaffold has been previously reported to possess antimicrobial properties. Epinephrine, despite forming fewer hydrogen bonds in docking studies, may achieve effective binding through optimal shape complementarity, as reflected by its lowest RMSD value. Additionally, catecholamines have been reported to influence bacterial physiology through interactions with bacterial adrenergic receptors, which may contribute to observed antimicrobial effects independently of β-lactamase inhibition.

The identification of epinephrine and omeprazole as potential β-lactamase inhibitor adjuvants carries significant clinical implications. Both compounds are widely available, inexpensive, and have well-established safety profiles spanning decades of clinical use. However, the concentrations required for antimicrobial potentiation (125–250 µg/mL) substantially exceed typical plasma concentrations achieved with standard dosing. Therefore, local delivery strategies, formulation modifications, or topical applications for urinary tract or wound infections may represent more feasible initial clinical applications. The binding characteristics of these compounds differ fundamentally from clinically approved inhibitors like avibactam, which form covalent bonds with Ser70. While covalent inhibitors typically exhibit superior potency, non-covalent inhibitors offer potential advantages including broader spectrum activity and reduced susceptibility to resistance mutations affecting covalent binding.

Several limitations of this study should be acknowledged. The in-vitro evaluation was limited to a single ESBL-producing E. coli clinical isolate, and although E. coli represents the most frequently encountered ESBL producer, validation across multiple species and clinical isolates is required. Additionally, agar-based diffusion methods provide qualitative evidence of synergistic interactions but do not permit calculation of precise MIC values or fractional inhibitory concentration indices (FICI) for quantitative synergy assessment. The mechanism of β-lactamase inhibition was not directly confirmed through enzyme kinetic studies, and the molecular genotype of the ESBL expressed by the clinical isolate was not determined by PCR. Future research should focus on evaluating activity against diverse ESBL-producing organisms including Klebsiella pneumoniae and Pseudomonas aeruginosa with molecularly characterized ESBL genotypes, performing standardized broth microdilution assays for precise MIC determination, conducting spectrophotometric enzyme inhibition assays to determine inhibition constants and characterize the inhibition mechanism, and exploring structural analogs to identify derivatives with improved potency. Following confirmation of broad-spectrum in-vitro activity, evaluation in relevant animal infection models would be warranted to advance these compounds toward potential clinical application.

Conclusion

This study employed a drug repurposing strategy integrating structure-based virtual screening, molecular dynamics simulations, and in-vitro antimicrobial assays to identify FDA-approved compounds capable of potentiating β-lactam antibiotics against ESBL-producing bacteria. From a library of 400 FDA-approved compounds, four candidates; epinephrine, omeprazole, sulfadimethoxine, and captopril were identified through molecular docking based on favorable binding affinities to SHV-1 β-lactamase. Comprehensive 250 ns molecular dynamics simulations validated the stability of all protein–ligand complexes, with the SHV-1/epinephrine complex demonstrating the lowest RMSD and most compact structure among all tested compounds.

In-vitro evaluation against an ESBL-producing Escherichia coli clinical isolate revealed that epinephrine and omeprazole effectively restored the antimicrobial activity of ceftriaxone and cefuroxime, with the ceftriaxone/omeprazole combination achieving the highest potentiation (20 mm zone diameter at 250 µg/mL). Notably, captopril showed no in-vitro activity despite favorable computational predictions, highlighting the essential role of experimental validation in drug discovery.

These findings demonstrate that FDA-approved non-β-lactam compounds can serve as β-lactamase inhibitor adjuvants, offering a rapid and cost-effective approach to combat ESBL-mediated antibiotic resistance. Epinephrine and omeprazole, in particular, emerge as promising candidates for further development given their established safety profiles, widespread availability, and low cost. Future studies focusing on expanded strain testing, enzyme kinetic characterization, and in-vivo efficacy evaluation are warranted to advance these compounds toward potential clinical application as antibiotic adjuvants.

Author contributions

NMO: Conceptualization, Methodology, Writing review & editing, Project administration; NMH: Conceptualization, Investigation, Methodology, Writing original draft , Writing review & editing , Formal analysis; SGL: Conceptualization, Methodology, Writing review & editing, Project administration; MSA: , Investigation, Methodology, Writing original draft ; SD: Investigation, Methodology, Writing original draft; MOS: , Investigation, Methodology, Writing original draft; AAJ: , Investigation, Methodology, Writing original draft ; MON: , Investigation, Methodology, Writing original draft; OFE: , Investigation, Methodology, Writing original draft ; SS: Investigation, Methodology, Writing original draft; TA: Writing review & editing , Formal analysis ; SAM: Writing review & editing , Visualization.

Funding

This research did not receive any external funding.

Data availability

The data sets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Clinical trial number

Not applicable.