Combating Campylobacter Resistance in Moroccan Poultry Using Essential Oil–Ciprofloxacin Combination

ABSTRACT

Background

Campylobacter is one of the leading causes of gastroenteritis, and the increasing prevalence of antimicrobial resistance calls for new treatment alternatives, such as essential oils (EOs).

Objectives

This study aims to investigate the prevalence of antimicrobial‐resistant Campylobacter strains recovered from poultry in Morocco, and to evaluate the antibacterial effects of the commercial EOs from cinnamon (CIEO), clove (CLEO) and oregano (OREO) alone and in combination with ciprofloxacin on Campylobacter coli and Campylobacter jejuni using in vitro and in silico approaches.

Methods

A total of 643 caecal samples were collected from poultry across four regions of Morocco, and C. coli and C. jejuni strains were isolated and tested for antibiotic susceptibility, following EUCAST guidelines. The antibacterial activity of selected EOs was assessed individually and in combination with antibiotics through broth microdilution and checkerboard assays. Additionally, molecular docking was conducted between the major EOs chemotypes and resistance‐associated protein targets (RNApα, CmeB, CmeR, DHPS, Erm(B), GyrA, GyrB, PBP2 and TetR).

Results

The prevalence of Campylobacter spp. among the collected samples was 15.52% (100/643). Many isolates exhibited strong resistance to nalidixic acid (90%), ciprofloxacin (85%) and tetracycline (78%). The studied EOs demonstrated a high antibacterial activity, with the MIC value ranging from 0.0312% to 0.0625% (v/v). A high level of synergism was achieved with OREO, while the interactions with CIEO and CLEO were considered antagonistic. In silico molecular docking revealed that caryophyllene exhibited strong binding affinities towards key resistance‐associated proteins, including GyrB (binding energy: −7.4 kcal/mol), CmeB (−7.5 kcal/mol) and PBP2 (−6.4 kcal/mol), suggesting potential interference with DNA gyrase activity, multidrug efflux and cell wall synthesis, respectively.

Conclusions

These findings suggest that selected EOs, particularly OREO, warrant further investigation as potential adjuncts to conventional antibiotics against multidrug‐resistant Campylobacter.

With an estimated prevalence of 15.52% in Moroccan poultry, Campylobacter species pose a significant food safety concern. We investigated whether essential oils (EOs) could restore the efficacy of ciprofloxacin against multidrug‐resistant (MDR) strains. The combination of ciprofloxacin with EOs produced a synergistic effect, significantly reducing the MICs of the antibiotic. Oregano oil emerged as the most effective synergist, substantially enhancing the antibacterial activity. This EO–antibiotic synergy provides a novel, natural strategy to combat Campylobacter resistance, potentially reducing the reliance on high‐dose antibiotics and mitigating the spread of resistance in agricultural settings

.

Abbreviations

AMR

antimicrobial resistance

CIP

ciprofloxacin

CYP

cytochrome P450

EOs

essential oils

ERY

erythromycin

GEN

gentamicin

HIA

human intestinal absorption

LogP

octanol–water partition coefficient

MALDI‐TOF

matrix‐assisted time of flight laser desorption mass spectrometry

MBC

determination of the minimum bactericidal concentration

mCCDA

modified cefoperazone, charcoal and deoxycholate

MDR

multidrug‐resistant

MICs

minimum inhibitory concentrations

MS

mass spectra

MW

molecular weight

NAL

nalidixic acid

nHA

number of hydrogen bond acceptors

nHD

number of hydrogen bond donors

P‐gp

P‐glycoprotein

RI

retention index

STR

streptomycin

TET

tetracycline

Vd

volume of distribution

βNAD

β‐nicotinamide adenine dinucleotide

1. Introduction

Campylobacter species, particularly Campylobacter jejuni and Campylobacter coli, are among the leading causes of bacterial gastroenteritis and are transmitted primarily through contaminated food (Dochania et al. 2026), especially poultry, which is considered a principal reservoir (Njoga et al. 2025). Although the overall clinical and public health significance of Campylobacter remains underestimated, particularly in developing countries such as Morocco, the considerable burden posed by this pathogen cannot be denied. This severity was highlighted by the European Food Safety Authority, which reported a significant increase in cases of campylobacteriosis, with 168,396 cases, more than double the number of salmonellosis infections (EFSA 2025). Furthermore, the spread of antimicrobial resistance within this pathogen is extremely concerning, as it compromises the efficacy of treatments, especially the first‐line therapies that have reached extremely high rates of resistance, notably macrolides and fluoroquinolones (EFSA and ECDC 2024), highlighting exactly why fluoroquinolone‐resistant Campylobacter has been classified as a high‐priority tier of the WHO global list for research and development (Tacconelli et al. 2018). Consequently, there is an urgent need to develop new therapeutic strategies to control the spread of multidrug‐resistant (MDR) pathogens, particularly Campylobacter.

Recently, essential oils (EOs) from plants have been highlighted as a promising solution due to their antimicrobial activity against highly resistant strains, whether used alone or as an adjunct to antibiotics (Raikwar et al. 2024; Rais et al. 2025). Interestingly, clove, cinnamon and oregano EOs are indeed recognised for their rich composition of key compounds, notably eugenol, thymol and cinnamaldehyde, known for their potent, broad‐spectrum antibacterial activities, particularly against antimicrobial‐resistant Campylobacter strains (Romanescu et al. 2023; Touhtouh et al. 2023).

Nevertheless, data regarding the activity of EOs against Campylobacter remain extremely limited. To our knowledge, only four studies have addressed this issue (Ahmed et al. 2016; Duarte et al. 2016; Gahamanyi et al. 2020; Kovács et al. 2016), highlighting that the number of EOs tested so far is still quite small (Seres‐Steinbach et al. 2026). Furthermore, only one study has examined their synergistic combinations with antibiotics, demonstrating a significant reduction in tetracycline and ampicillin minimum inhibitory concentrations (MICs), alongside the disruption of biofilm formation (El Baaboua et al. 2022). While this approach offers considerable advantages (Alkufeidy et al. 2024), data regarding interaction with first‐line antibiotics, particularly ciprofloxacin, remain absent.

Ultimately, not only is there a general lack of promising treatment studies for this pathogen, but specific studies focusing on the molecular interactions between EO chemotypes and Campylobacter resistance targets also remain largely unexplored. To address this gap, in silico approaches like molecular docking are highly valuable. By simulating these exact targets, allowing the prediction of binding affinities and elucidating the antimicrobial mechanism of EOs compounds (Yuan et al. 2023).

Considering this, the present study aimed to (i) determine the chemical composition of the tested EOs; (ii) examine antibiotic resistance in strains of avian Campylobacter spp. collected from different regions of Morocco; (iii) evaluate the antibacterial activities of CIEO, CLEO and OREO, alone as well as in combination with ciprofloxacin, on Campylobacter spp. strains resistant to ciprofloxacin; and (iv) investigate the binding affinities between the major EOs components and key resistance‐related targets in C. coli and C. jejuni.

2. Materials and Methods

2.1. Ethical Considerations

The study used post mortem samples from animals destined for consumption, in accordance with Moroccan food safety law 28‐07. Since there was no live animal involvement, IACUC approval was not required. Collection and handling were carried out by standard ISO 10272‐1:2017 to avoid cross‐contamination. The approach aligns with the 3Rs principle of ethically using by‐products from the food industry.

2.2. Sample Collection

A total of 643 caecal samples were collected from Turkeys (n = 294) and chickens (n = 349) across four regions of Morocco: the western region (R1, n = 139), the northwestern region (R2, n = 127), the northern and northeastern region (R3, n = 354) and the central‐eastern region (R4, n = 23). Sampling was carried out on a convenience basis depending on bird availability on each visit, between January and December 2023. The four Slaughterhouses were selected based on their status as the major facilities supplying most of the poultry production within each region.

All caeca were obtained from clinically healthy animals aged 6–8 weeks. Samples were collected under aseptic conditions immediately after slaughter, transported to the laboratory and processed the same day.

2.3. Isolation and Identification of Campylobacter Species

Campylobacter species were isolated following the standard (ISO 10272‐1 2017). Direct inoculation on the plate surface of modified cefoperazone, charcoal and deoxycholate (mCCDA) medium (Oxoid, Cambridge, UK), followed by incubation at 42°C for 48 h in a microaerophilic condition. For the purification step, suspected Campylobacter were grown on 5% horse blood Columbia agar plates (Oxoid, Cambridge, UK) and incubated under the same conditions. The pure colonies were placed in a suspension of 0.85% sterile saline, placed on a microscope slide and Gram‐stained. The Gram‐negative, curved isolates were subjected to oxidase, catalase, indoxyl acetate and hippurate hydrolysis tests. Additional analyses included growth at 25°C under microaerobic conditions, as well as urease activity, glucose utilisation and hydrogen sulphide (H₂S) production.

2.4. Confirmation by MALDI‐TOF

The identification of Campylobacter isolates was carried out using MALDI‐TOF, by placing colonies into a disposable target plate (Bruker Daltonics, Bremen, Germany), and then coated with a matrix solution (1 µL) containing a‐cyano‐4‐hydroxycinnamic acid (Bruker Daltonics, Bremen, Germany), then inserted in the MALDI Biotyper mass spectrometer, MBT smart (Bruker Daltonics, Bremen, Germany), and the peptidic spectra obtained were matched with the extensive Bruker MALDI Biotyper library (version 5627) and software (version 3.4; Assi et al. 2025).

2.5. Statistical Analysis

A statistical analysis was carried out in Python (v 3.14.0), with the Pandas library (v 2.3.3) for data management, SciPy (v 1.16.3) for statistical testing and NumPy (v 2.3.4) for numerical calculations. To assess whether the positivity rate of Campylobacter was associated with sampling region, a Pearson's chi‐square (χ ²) test of independence was performed using the number of positive versus negative samples by region. Cramér's V was calculated to measure the strength of the association. Statistical significance was set at α = 0.05, with p < 0.05 considered statistically significant.

2.6. Antimicrobial Susceptibility Testing

All Campylobacter isolates were tested for antimicrobial susceptibility towards six antimicrobials using the Sensititre EUCAMP2 plate (Thermo Fisher Scientific, East Grinstead, UK), including the following concentration ranges: erythromycin (1–128 µg/mL), ciprofloxacin (0.12–16 µg/mL), tetracycline (0.5–64 µg/mL), streptomycin (0.25–16 µg/mL), nalidixic acid (1–64 µg/mL) and gentamicin (0.12–16 µg/mL). An inoculum suspension (100 µL) with 0.5 McFarland adjustment was mixed with 11 mL of cation‐adjusted Mueller–Hinton broth (Oxoid, Thermo Fisher Scientific, Basingstoke, UK). Then, 50 µL of the suspension was dispensed into each well of a 96‐well microplate using an automated dispenser (Thermo Fisher Scientific, Basingstoke, UK). The microplate was then sealed and incubated for 24 h at 42°C under microaerophilic conditions using a GENbox microaer system (bioMérieux, Marcyl l'Etoile, France) with an O₂ concentration between 6.2% and 13.2% and a CO₂ concentration between 2.5% and 9.5%. Results were interpreted according to the EUCAST (v.1.0; Amara et al. 2024) and the European Union Reference Laboratory for Antimicrobial Resistance. Resistance cut‐offs for C. jejuni and C. coli, respectively, were: erythromycin (> 4 µg/mL; > 8 µg/mL), tetracycline (> 1 µg/mL; > 2 µg/mL), and for both species: ciprofloxacin (> 0.5 µg/mL), streptomycin (> 4 µg/mL), nalidixic acid (> 16 µg/mL) and gentamicin (> 2 µg/mL). C. jejuni ATCC 33560 was used for quality control.

2.7. Antimicrobial Susceptibility Testing of Essential Oils on Campylobacter spp.

2.7.1. Essential Oils

The CIEO (batch No. 32401) and CLEO (batch No. 44101) were obtained from NatureEsoin, whereas OREO was purchased from the APIA brand. The manufacturer did not disclose the exact botanical species, the precise extraction protocol or the geographic origin of the plant‐based raw material. Furthermore, the batch number was not indicated for the oregano EO, a common omission in commercially available EOs not intended for research, although the product was labelled as 100% pure. The chemical composition is therefore verified by GC‐M analysis.

The EOs were stored in their original sealed container at 4°C, protected from light, until chemical and biological analyses were conducted.

2.7.2. Gas Chromatography–Tandem Mass Spectrometry Analysis

CIEO, CLEO and OREO were analysed using gas chromatography–tandem mass spectrometry (GC–MS/MS) as described in our previous study (Benali et al. 2024). The procedure used a triple quadrupole tandem mass spectrometer linked to a gas chromatography TQ8040 NX (Shimadzu, Tokyo, Japan) system. An Rxi‐5Sil MS non‐polar capillary column (30 m × 0.25 mm ID × 0.25 µm) was used. The carrier gas used was pure helium, and the injection volume was 1 µL. The source temperature was 200°C while the chromatographic system was set up for splitless injections, which split opening at 4 min, a 250°C injector temperature, and 37.1 kPa injector pressure. In this order, the temperature was planned to rise from 50°C to 160°C for 2 min, then to 280°C for 2 min.

The identification of each product was based on the comparison of its retention index (RI), which calculated using n‐alkanes series between C₈ and C₂₀, and its mass spectra (MS) spectra with those documented in the literature, and by computer matching with standard reference databases (NIST2019).

2.8. Agar‐Well Diffusion Assay

This step was set up as a preliminary qualitative screening assay to assess the presence or absence of antimicrobial activity of EOs against C. coli and C. jejuni. The method employed is based on the use of Mueller–Hinton agar medium (Condalab, Madrid, Spain) mixed with 5% lysed horse blood and β‐nicotinamide adenine dinucleotide (β‐NAD; Thermo Fisher Scientific, Heysham Lancashire, UK). The Petri dishes were inoculated with an inoculum suspension adjusted to the 0.5 Macfarland standard. A well with a diameter of 6.0 mm was then drilled into the surface of each plate. Then, 50 µL of each EO (undiluted) was dispensed into the corresponding wells, and the Petri plates were incubated in microaerophilic conditions at 42°C for 24 h. Control plates (without EOs) were incubated under the same conditions.

2.9. Determining the Minimum Inhibitory Concentration

The MIC of the EOs was determined against eleven MDR Campylobacter strains. Briefly, a 96‐well microplate was prepared with 100 µL of Bolton's broth medium, and 100 µL of 4% EO dilution (prepared by diluting the EO in phosphate‐buffered saline (PBS) containing 1.0% (v/v) Tween 80; Y. Ge and Ge 2016), which was added to the first column of the microplate, from which a series of 1:2 dilutions was made until a concentration of 0.0078 (v/v) was reached. Finally, for each isolate, 10 µL of the bacterial suspension was added to the corresponding wells. All experiments were carried out in triplicate, accompanied by both positive growth control and negative control of the diluent used. The microplate set was incubated under microaerophilic conditions using a GENbox microaer system (bioMérieux, Marcyl l'Etoile, France) at 42°C for 24 h (Eloff 1998). The MIC values were determined by adding 10 µL of resazurin (0.01% w/v), followed by a 2 h incubation. The absence of growth is indicated by a purple or blue dye, while the pink dye means positive growth (El Baaboua et al. 2022). Any ambiguity encountered when visually interpreting the MIC values was clarified by applying the same methodology below (minimum bactericidal concentration [MBC] determination) to confirm the presence or absence of viability.

2.10. Determination of the Minimum Bactericidal Concentration

The wells from each concentration of EOs showed no visible growth were selected for assessment of bacterial viability, 10 µL of the content of each well was transferred onto Müller–Hinton agar plates (Candalab, Madrid, Spain) containing 5% lysed horse blood and β‐NAD (Thermo Fisher Scientific, Heysham Lancashire, UK) and incubated at 42°C for 24 h under microaerophilic conditions using a GENbox microaer system (bioMérieux, Marcyl l'Etoile, France) to check for any surviving bacterial colonies. The lowest concentration that showed no visible growth was recorded as the MBC. This indicates a total destruction of the bacterial population at that concentration.

2.11. Determination of the Interaction Between EOs and Antibiotics

2.11.1. Choice of Antibiotics

The selection of antibiotics was guided by previously reported resistance patterns, with ciprofloxacin commonly described as exhibiting high levels of resistance (Papoula‐Pereira et al. 2025). To further evaluate antimicrobial interactions, a series of concentrations was tested both individually and in combination with a selected dilution of the three EOs.

2.11.2. Checkerboard Microdilution Assays

The synergistic interaction between EOs and antibiotics against the same isolates used for MIC determination was investigated using the checkerboard method. The procedure was implemented in 96‐well microplates and involved the sequential preparation of three separate plates. First, the antibiotic plate was prepared by dispensing 50 µL of Bolton Broth (Oxoid, Cambridge, UK) into all wells (Columns 1–8). A 64 mg/mL antibiotic solution was then prepared, and 50 µL of this was added to the wells in Column 1, followed by serial dilution to Column 8. Next, to prepare the EOs’ plate, 100 µL was dispensed into each well. Next, 100 µL of the EOs stock solution (4% for CIEO, 2% for both CLEO and OREO) was added to the wells in the first row (A), followed by serial dilution to the eighth row (H). Finally, the Checkerboard plate was assembled by dispensing 50 µL of medium into the wells of the first row (A) and the first column (1). Next, 50 µL from the wells of the antibiotic plate were transferred from Column 1 to the corresponding wells of the Checkerboard plate, starting with Column 2. Similarly, 50 µL from the wells of the EO plate were transferred from row B to the corresponding wells of the Checkerboard plate. Once the plate setup was complete, an inoculum with a turbidity of 0.5 McFarland was prepared from each Campylobacter isolate, diluted 1:10 in Bolton broth, and 10 µL was inoculated into all wells to achieve a final concentration of 5 × 10⁵ CFU/mL per well (Table S1). The interaction between EOs and antibiotics was evaluated by calculating the fractional inhibitory concentration index (FICi), computed using the formula: FLCi = FIC(A) + FIC(B), where the FIC of each agent corresponds to the ratio of its combined MIC to its MIC when used alone. The interaction was classified as synergistic (FICi ≤ 0.5), partially synergistic (0.5 < FICi ≤ 0.75), additive (0.76 ≤ FICi ≤ 1), indifferent or non‐differential (1 < FICi ≤ 2) or antagonistic (FICi > 2; El Baaboua et al. 2022).

2.12. In Silico Analysis

2.12.1. Molecular Docking

Molecular docking analyses were conducted to complement our in vitro results. This computational approach allowed us to predict how the major compounds identified by GC–MS/MS in our EOs interact with proteins involved in antimicrobial resistance.

The three‐dimensional structures of RNA polymerase α‐subunit (RNApα), CmeB, CmeR, DHPS, Erm(B), GyrA, GyrB, PBP2 and TetR from C. jejuni and C. coli were selected as docking targets. Protein structures were obtained from the Protein Data Bank (PDB) or the AlphaFold database. For C. jejuni, the structures of RNApα (PDB ID: 4NOI), PBP2 (PDB ID: 8YJX), CmeR (PDB ID: 2QCO) and DHPS (PDB ID: 4LY8) were obtained from the PDB. The remaining targets, including CmeB (UniProt ID: A0A6H6KA52; mean pLDDT: 92.25), GyrB (O87667, mean pLDDT: 89), Erm(B) (A0A0C4M7A4; mean pLDDT: 93.12) and TetR (E1CJK9; mean pLDDT: 90.88), were retrieved from the AlphaFold database. For C. coli, the structures of GyrA (A0A5T0ITB8; mean pLDDT: 86.44), CmeB (Q1A4H5; mean pLDDT: 91.69), GyrB (A0A5T1ZIF3; mean pLDDT: 86.38), PBP2 (A0A5T1TET3), Erm(B) (A0A0S3CVF4; mean pLDDT: 93.31), TetR (A0A2S0DAX2; mean pLDDT: 78.94), CmeR (A0A1L2IW69; mean pLDDT: 90.38) and DHPS (A0A6C7JZ85; mean pLDDT: 98.19) were obtained from the AlphaFold database. The pLDDT values indicated high predicted structural confidence for the AlphaFold models used in the docking analysis.

Additionally, as no crystal structure was available for C. jejuni DNA gyrase, a homologous structure from E. coli (PDB ID: 6RKS) was selected due to its 82.50% sequence similarity within the quinolone resistance‐determining region (QRDR) of GyrA. To better represent C. jejuni, a site‐directed mutation was introduced using the Crystallographic Object‐Oriented Toolkit (WinCoot‐0.9.8.92; Paul Emsley), where Ser83 in E. coli GyrA was substituted with Thr to correspond to Thr86 in C. jejuni (Miura‐Ajima et al. 2024).

These targets were selected based on their established roles in Campylobacter resistance and physiology. RNApα, crucial for transcriptional regulation (Zakharova et al. 1998), was included as an essential survival target. The CmeABC efflux pump system contributes significantly to broad‐spectrum antibiotic resistance; therefore, both the CmeB transporter and its transcriptional repressor CmeR were selected (Lin et al. 2002, 2005). DNA gyrase, composed of GyrA and GyrB subunits, is essential for DNA supercoiling and represents the primary target of fluoroquinolones; resistance primarily results from mutations in gyrA, notably the Thr‐86‐Ile substitution (Champoux 2001; B. Ge et al. 2005; Han et al. 2012). Penicillin‐binding protein 2 (PBP2), a key enzyme in peptidoglycan cross‐linking, was chosen for its role in bacterial cell wall integrity (Choi et al. 2024). Dihydropteroate synthase (DHPS), essential in folate biosynthesis, was included as a validated antimicrobial target (Gibreel and Sköld 1999). Erm(B), which confers macrolide resistance through methylation of the 23S rRNA peptidyl transferase region at positions 2058 and 2059 (Vacher et al. 2003), and TetR, a transcriptional regulator involved in efflux‐mediated resistance (Ramos et al. 2005), were incorporated to represent diverse resistance mechanisms.

Ligands were selected from the major constituents identified by GC–MS/MS with a relative abundance ≥ 5%. Four trace constituents (< 1%) were excluded from the docking analysis. All ligand structures were retrieved from the PubChem database. Prior to docking, drug‐likeness was assessed using Lipinski's rule of five; however, all major constituents were docked regardless of Lipinski compliance to ensure comprehensive screening, and ADME analysis was applied post‐docking to prioritise lead candidates. Ligand‐binding site predictions were carried out using the PrankWeb and DeepSite machine learning servers, which combine structural and sequence information to identify potential binding pockets (Jiménez et al. 2017; Krivák and Hoksza 2018). These tools also facilitate the detection of novel binding sites and can analyse multidomain protein structures (Touhtouh et al. 2025), which help to determine binding site coordinates and identify key amino acids involved. Besides, all the structures of the ligands were obtained from PubChem database. Molecular docking simulations were conducted by AutoDock Vina, while ligand preparation, charge distribution and non‐polar hydrogen merging were executed with AutoDockTools‐1.5.7 (Chraa et al. 2025; Touhtouh et al. 2025). The grid box parameters for each target protein were defined to encompass the binding site. The detailed grid box specifications for C. coli and C. jejuni targets are summarised in Table 1.

TABLE 1.

Grid box centre coordinates and dimensions for targeted proteins.

Species	Target	Structure ID	Source	Centre X (Å)	Centre Y (Å)	Centre Z (Å)	Grid box size
C. coli	CmeB	Q1A4H5	UniProt	7.9452	−1.8168	−10.2582	40 × 40 × 40
	GyrA	6RKW	UniProt	128.8694	158.6329	152.7551	40 × 40 × 40
	GyrB	A0A5T1ZIF3	UniProt	8.8707	−10.7802	−26.1881	40 × 40 × 40
	PBP	A0A5T1TET3	UniProt	−4.4520	−0.2777	26.5535	40 × 40 × 40
	Erm(B)	A0A0S3CVF4	UniProt	−10.8654	−4.5687	8.0157	40 × 40 × 40
	TetR	A0A2S0DAX2	UniProt	−1.9859	−5.8749	7.2185	40 × 40 × 40
	CmeR	A0A1L2IW69	UniProt	2.9120	−3.5860	−5.9680	40 × 40 × 40
	DHDPS	A0A5T1TET3	UniProt	1.8364	−9.5607	6.2222	40 × 40 × 40
C. jejuni	RNA Polymerase (β‐subunit)	4NOI	PDB	11.0070	11.3610	14.7590	40 × 40 × 40
	CmeB	A0A6H6KA52	AlphaFold	8.1336	−2.2696	−11.0188	40 × 40 × 40
	GyrA	Q6J8I1	UniProt	−30.9474	3.9422	27.8466	40 × 40 × 40
	GyrB	O87667	UniProt	5.0226	−9.5575	−29.4449	40 × 40 × 40
	PBP	8YJX	PDB	21.4476	−50.4299	−70.8348	40 × 40 × 40
	Erm(B)	A0A0C4M7A4	UniProt	−11.8059	−3.8633	7.4590	40 × 40 × 40
	TetR	E1CJK9	UniProt	2.7657	−0.2688	−0.2328	40 × 40 × 40
	CmeR	2QCO	PDB	35.6601	18.8423	41.3417	40 × 40 × 40
	DHDPS	4LY8	PDB	−24.8710	−4.0480	50.8980	40 × 58 × 70

The docking protocol was validated by re‐docking co‐crystallised ligands into their respective native binding sites. The root‐mean‐square deviation (RMSD) between the best re‐docked pose and the crystallographic pose was calculated; an RMSD < 3 Å confirmed the reliability of the docking parameters. For targets without available co‐crystallised ligands, the results are interpreted with appropriate caution. The final docking complexes were visualised and analysed using Discovery Studio 2024 Client.

2.12.2. Drug‐Likeness Studies

To assess the bioavailability of the identified phytochemicals, SWISS ADME (http://www.swissadme.ch/) was used. This platform investigates the drug potential of compounds by predicting their bioavailability based on Lipinski's rule of five. The prediction of oral bioavailability for compounds was achieved through an integrated analysis of parameters such as molecular weight (MW), octanol‐water partition coefficient (LogP), and the number of hydrogen bond acceptors (nHA) and donors (nHD; Kaushik et al. 2014). Compounds that fulfilled the criteria—MW ≤ 500, LogP ≤ 5, nHA ≤ 10 and nHD ≤ 5—were chosen for docking studies (Xiong et al. 2021).

2.12.3. Pharmacokinetic Study

To comprehensively evaluate the pharmacokinetic and toxicity profiles of the compounds, an in silico ADMET analysis was conducted using the pkCSM web server (http://biosig.unimelb.edu.au/pkcsm/; Ashraf et al. 2021). Absorption was assessed through hydrophilicity, Caco‐2 permeability and interaction with P‐glycoprotein (P‐gp). Distribution parameters included volume of distribution (Vd), plasma‐free fraction and blood–brain barrier permeability. Metabolism was analysed based on interactions with cytochrome P450 enzymes, while excretion was evaluated in terms of total clearance and OCT2 transporter involvement. Toxicity predictions included outcomes from the AMES test, as well as potential hepatotoxicity and skin toxicity (Pires et al. 2015).

3. Results

3.1. Occurrence of the Campylobacter Species

According to biochemical confirmation and MALDI‐TOF analysis, a total of 100 (15.55%) Campylobacter isolates were recovered from 643 samples. Among these, 90 were classified as C. coli and 10 as C. jejuni (Figure 1).

FIGURE 1.

Proportion of identified Campylobacter spp.

Regarding the variation in distribution across the sampled regions, 14.1% of isolates were recorded in the western region (R1), 20.1% in the northwestern region (R2), 15.7% in the northern and northeastern region (R3) and finally, 8.7% of isolates were detected in the central‐eastern region (R4). Table 2 listed the number of samples and isolates collected by region. Based on the statistical analysis, the chi‐square test of independence showed no significant association (p > 0.05) between sampling region and overall Campylobacter positivity (χ ² (3, N = 643) = 3.61, p = 0.31, Cramer's V = 0.08), indicating similar positivity rates across the four studied regions (Figure 2).

TABLE 2.

Regional distribution of collected samples, overall positivity and species isolation.

Sampled region	Number of samples	Number of positive samples	C. coli	C. jejuni
R1	139	28	26	2
R2	127	20	19	1
R3	354	50	44	6
R4	24	2	1	1
Total	644	100	90	10

FIGURE 2.

Distribution of Campylobacter spp. in four different regions.

3.2. Evaluation of the Antimicrobial Susceptibility Profile of Campylobacter Species

Campylobacter strains confirmed were tested for susceptibility against six antimicrobials using Thermo Fisher Sensititre MIC plates. However, EUCAST breakpoints are only available for four of these antibiotics: ciprofloxacin (CIP), tetracycline (TET), erythromycin (ERY) and gentamicin (GEN). Therefore, the interpretation of results for nalidixic acid (NAL) and streptomycin (STR) was based on the guidelines provided by the European Union Reference Laboratory for Antimicrobial Resistance. Notably, 90% of Campylobacter isolates tested were resistant to nalidixic acid, 85% to ciprofloxacin, 78% to tetracycline, 47% to erythromycin and 36% to streptomycin; however, no isolates were resistant to gentamicin.

Analysis of the intersectionality between antimicrobial resistance profiles revealed significant overlap. The most prominent intersection comprised 24 isolates resistant to all antimicrobials tested except gentamicin. In stark contrast, only eight isolates were susceptible. By definition, multidrug resistance (MDR) is the acquired resistance to at least one agent from three or more classes of antimicrobials. Analysis revealed that 55% of the strains are MDR (including 49% of C. coli and 6% of C. jejuni), 25% are resistant to two classes and 12% are resistant to just one class of antimicrobial (Figure 3).

FIGURE 3.

Antimicrobial resistance overlaps among Campylobacter isolates.

3.3. Chemical Composition of EOs

The GC/MS/MS analysis revealed that five compounds were identified in CIEO and OREO, and four in CLEO (Table 3). cis‐Cinnamaldehyde (85.62%) and eugenol (82.11%) were the major identified components of CIEO and CLEO, respectively, while thymol (77.8%) was the main compound of OREO.

TABLE 3.

Chemical composition of CIEO, OREO and CLEO.

EOs		Compounds	RI a	Area %
CIEO	1	2‐Methoxybenzaldehyde	1569	1.02
	2	cis‐Cinnamaldehyde	1580	85.62
	3	Cinnamyl acetate	1815	1.56
	4	2‐Propenoic acid	1824	3.33
	5	(Z)‐2‐Methoxycinnamaldehyde	1871	8.12
		Total		99.65
OREO	1	β‐Linalool	1521	0.99
	2	Terpinen‐4‐ol	1550	0.46
	3	Carvacrol	1581	19.62
	4	Thymol	1584	77.8
	5	Caryophyllene	1803	1.13
		Total		100
CLEO	1	Eugenol	1625	82.11
	2	Caryophyllene	1800	4.57
	3	Humulene	1804	0.73
	4	Eugenol acetate	1808	12.23
		Total		99.64

^a RI: identification by retention index relative to C₈–C₂₀ on Rxi‐5 Sil MS.column.

3.4. Agar‐Well Diffusion Assay

The preliminary antimicrobial activity of OREO, CIEO and CLEO against pure C. coli and C. jejuni isolates was assessed using an agar‐well diffusion assay. This initial screening aimed to identify EOs with inhibitory potential before quantitative analysis. The results demonstrated a clear zone of complete growth inhibition around wells containing each EO for both bacterial isolates. In contrast, Bacterial colonies on control plates (without EOs) exhibited confluent growth. This finding confirmed the antimicrobial efficacy of the tested EOs and underscored the relevance of moving on to the next stage to quantify more precisely the inhibitory concentration of each EO.

3.5. Determination of MIC and MBC of EOs

Following the promising results from the initial screening, the MIC of the three EOs was determined. The results demonstrated a high susceptibility of the tested isolates with MIC values ranging from 0.0625% to 0.0312% (v/v) for all EOs.

To further characterise the nature of the antimicrobial activity, the MBC was evaluated by subculturing from the MIC test wells. The MBC was defined as the lowest concentration resulting in no growth. For all three EOs, the MBC values were found to be identical to their respective MIC values, yielding an MBC/MIC ratio of 1 ≤ 2. The activity of OREO, CIEO and CLEO is definitely classified as bactericidal against both C. coli and C. jejuni isolates Table 4.

TABLE 4.

MIC and MBC values of cinnamon, clove and oregano EOs against ciprofloxacin‐resistant Campylobacter spp.

	CIEO			OREO			CLEO
EOs strains	MIC	MBC	Effect	MIC	MBC	Effect	MIC	MBC	Effect
C. jejuni	0.0625	0.0625	Bactericide	0.0625	0.0625	Bactericide	0.0625	0.0625	Bactericide
C. jejuni	0.0312	0.0312	Bactericide	0.0312	0.0312	Bactericide	0.0312	0.0312	Bactericide
C. jejuni	0.0312	0.0312	Bactericide	0.0312	0.0312	Bactericide	0.0312	0.0312	Bactericide
C. jejuni	0.0625	0.0625	Bactericide	0.0625	0.0625	Bactericide	0.0625	0.0625	Bactericide
C. jejuni	0.0625	0.0625	Bactericide	0.0625	0.0625	Bactericide	0.0625	0.0625	Bactericide
C. coli	0.0312	0.0312	Bactericide	0.0312	0.0312	Bactericide	0.0312	0.0312	Bactericide
C. coli	0.0312	0.0312	Bactericide	0.0312	0.0312	Bactericide	0.0312	0.0312	Bactericide
C. coli	0.0312	0.0312	Bactericide	0.0312	0.0312	Bactericide	0.0312	0.0312	Bactericide
C. coli	0.0625	0.0625	Bactericide	0.0625	0.0625	Bactericide	0.0625	0.0625	Bactericide
C. coli	0.0625	0.0625	Bactericide	0.0625	0.0625	Bactericide	0.0625	0.0625	Bactericide
C. coli	0.0312	0.0312	Bactericide	0.0312	0.0312	Bactericide	0.0312	0.0312	Bactericide

3.6. Combined Effect of Ciprofloxacin and EOs

The interactions between the three EOs and ciprofloxacin were investigated using the checkerboard microdilution method against 11 resistant strains of Campylobacter spp., and the FICi was determined. The combination of OREO and CIP demonstrated a decrease in the antibiotic's MIC against all the studied strains, which went from 8 µg/mL to 0.25 µg/mL in the presence of EO. This is illustrated by the FICI values, which were all below 0.5, indicating a synergistic effect of OREO when combined with CIP against Campylobacter spp. resistant strains (Table 5).

TABLE 5.

The effect of combining EOs studied with CIP on Campylobacter spp.

	CIP (µg/mL) and CLEO (%)					CIP (µg/mL) and CIEO (%)					CIP (µg/mL) and OREO (%)
Strains	MIC of CIP	MIC of CIP in the presence of CLEO	MIC of CLEO	MIC of CLEO in the presence of CIP	FIC	MIC of CIP	MIC of CIP in the presence of CIEO	MIC of CIEO	MIC of CIEO in the presence of CIP	FIC	MIC of CIP	MIC of CIP in the presence of OREO	MIC of OREO	MIC of OREO in the presence of CIP	FIC
C. jejuni C. jejuni	8	16	0.0625	0.0625	3	8	16	0.0312	0.0312	3	8	≤ 0.25	0.0625	0.0156	0.2808
C. jejuni	8	16	0.0312	0.0625	2.4992	8	16	0.0625	0.0625	3	8	≤ 0.25	0.0625	0.0156	0.2808
C. jejuni	8	16	0.0312	0.0625	2.4992	8	16	0.0625	0.0625	3	8	≤ 0.25	0.0625	0.0156	0.2808
C. jejuni	8	16	0.0625	0.0625	3	8	16	0.0312	0.0625	2.4992	8	≤ 0.25	0.0625	0.0156	0.2808
C. coli	8	16	0.0625	0.0625	3	8	16	0.0625	0.0625	3	8	≤ 0.25	0.0625	0.0156	0.2808
C. coli	8	16	0.0312	0.0625	2.4992	8	16	0.0625	0.0625	3	8	≤ 0.25	0.0312	≤ 0.0078	0.2812
C. coli	8	16	0.0312	0.0625	2.4992	8	16	0.0625	0.0625	3	8	≤ 0.25	0.0312	≤ 0.0078	0.2812
C. coli	8	16	0.0625	0.0625	3	8	16	0.0312	0.0312	3	8	≤ 0.25	0.0312	≤ 0.0078	0.2812
C. coli	8	16	0.0625	0.0625	3	8	16	0.0312	0.0312	3	8	≤ 0.25	0.0312	≤ 0.0078 ≤ 0.0078	0.2812
C. coli	8	16	0.0625	0.0625	3	8	16	0.0312	0.0625	2.4992	8	≤ 0.25	0.0312	≤ 0.0078	0.2812
	8	16	0.0625	0.0625	3	8	16	0.0312	0.0312	3	8	≤ 0.25	0.0625		0.1508

Conversely, combinations of CIP with either CLEO or CIEO resulted in an antagonistic interaction, as indicated by FICI values exceeding 2. As a result, the MIC of CIP mainly increased from 8 to 16 µg/mL (Table 5).

3.7. In Silico Analysis

To provide a comprehensive understanding of the antimicrobial potential, we further investigated the EOs' mechanisms of action through chemical characterisation via GC–MS–MS, followed by in silico predictions including drug‐likeness assessment, ADMET profiling and molecular docking. This approach allowed us to identify key bioactive compounds and their interactions with bacterial targets.

3.8. Drug‐Likeness Assessments

GC–MS–MS analysis allowed us to detect 13 volatile compounds in the three EOs. Nine of these compounds were selected for further analysis, while four compounds present in traces (< 1%) were excluded. Drug‐likeness evaluation of the selected compounds revealed that all nine satisfied Lipinski's rules (Table 6). Based on these favourable results, all nine compounds were subjected to molecular docking studies.

TABLE 6.

Drug‐likeness evaluation of phytoconstituents.

Molecules	MW (g/mol)	HD	HA	XlogP3‐AA	Good profile
2‐Methoxybenzaldehyde	136.15	0	2	1.7	Yes
2‐Propenoic acid	72.06	1	2	0.3	Yes
(Z)‐2‐Methoxycinnamaldehyde	162.18	0	2	1.7	Yes
Carvacrol	150.22	1	1	3.1	Yes
Thymol	150.22	1	1	3.3	Yes
Eugenol	164.20	1	2	2	Yes
Caryophyllene	204.35	0	0	4.4	Yes
Eugenol acetate	206.24	0	3	2.3	Yes
cis‐Cinnamaldehyde	132.16	0	1	1.9	Yes

Abbreviations: HA, hydrogen bond acceptor count; HD, hydrogen bond donor count; MW, molecular weight.

3.9. Molecular Docking

Molecular docking was performed to investigate potential interactions between the major EO constituents and selected resistance‐associated proteins of C. jejuni and C. coli. The selected targets encompass mechanisms critical to antimicrobial resistance and bacterial survival, including DNA replication and supercoiling (GyrA, GyrB, RNApα), multidrug efflux (CmeB, CmeR, TetR), cell wall synthesis (PBP2), folate biosynthesis (DHPS) and macrolide resistance (ErmB).

According to Table 7, carvacrol, thymol, eugenol, caryophyllene and eugenol acetate exhibited the most favourable binding energies across the C. jejuni targets, while the remaining compounds displayed comparatively weaker scores. For RNApα, thymol exhibited a binding energy of −6.7 kcal/mol, compared to the standard antibiotics, which scored between −6.0 and −6.5 kcal/mol. Caryophyllene demonstrated favourable binding energies towards CmeB (−7.5 kcal/mol), CmeR (−7.5 kcal/mol), DHDPS (−7.0 kcal/mol), GyrA (−5.9 kcal/mol), GyrB (−7.4 kcal/mol) and TetR (−7.1 kcal/mol), while the reference antibiotics ranged from −5.4 to −8.6 kcal/mol across these targets (Figure 4). The dominant interaction types included van der Waals forces (CmeB), π–π stacking (CmeR, GyrA, GyrB) and π–alkyl interactions (DHDPS), while π–sigma bonds were predominant in TetR. Carvacrol showed a binding energy of −6.1 kcal/mol with ErmB. For PBP2, eugenol acetate displayed a binding energy of −6.6 kcal/mol, forming two conventional hydrogen bonds (HIS554 and GLU421), a π–cation interaction (HIS576) and a π–alkyl interaction (Figure 5).

TABLE 7.

Docking results of main chemicals towards resistance‐related targets (kcal/mol).

	Molecules	RNApα	CmeB	CmeR	DHDPS	ErmB	GyrA	GyrB	PBP2	TetR
C. jejuni	2‐Methoxybenzaldehyde	−4.5	−5.3	−4.9	−5.3	−5.2	−3.9	−5.0	−5.4	−5.0
	2‐Propenoic acid	−4.3	−3.7	−3.6	−4.0	−4.2	−3.0	−3.8	−3.5	−3.5
	(Z)‐2‐Methoxycinnamaldehyde	−4.7	−5.8	−5.3	−5.6	−4.6	−4.4	−5.7	−5.9	−5.3
	Carvacrol	−5.5	−6.1	−5.8	−6.2	−6.1	−4.6	−5.9	−6.1	−5.7
	Thymol	−6.7	−5.8	−5.7	−6.3	−5.7	−5.0	−5.6	−5.9	−5.9
	Eugenol	−4.6	−6.1	−5.6	−6.9	−5.4	−4.6	−6.0	−6.0	−5.6
	Caryophyllene	−5.7	−7.5	−7.5	−7.0	−5.5	−5.9	−7.4	−6.4	−7.1
	Eugenol acetate	−5.1	−6.7	−6.1	−6.8	−5.2	−4.4	−6.3	−6.6	−6.1
	cis‐Cinnamaldehyde	−4.5	−5.4	−5.3	−5.6	−5.4	−4.3	−5.4	−5.3	−5.2
Standards	Ampicillin	−6.5	−8.2	−8.2	−8.2	−5.8	−5.7	−6.7	7.8	−7.9
	Erythromycin	−6.4	−7.8	−6.9	−6.8	−6.0	−5.4	−5.4	−6.6	−8.6
	Ciprofloxacin	−6.0	−8.2	−8.1	−7.7	−6.3	−5.7	−7.5	−7.5	−8.0
C. coli	2‐Methoxybenzaldehyde	—	−5.3	−4.9	−5.4	−4.7	−5.0	−5.1	−4.4	−3.4
	2‐Propenoic acid	—	−3.9	−3.5	−3.8	−4.2	−3.6	−3.8	−3.1	−2.4
	(Z)‐2‐Methoxycinnamaldehyde	—	−5.8	−5.4	−5.2	−5.9	−5.9	−5.7	−4.4	−3.8
	Carvacrol	—	−5.7	−5.7	−5.9	−6.4	−6.2	−5.8	−4.9	−4.3
	Thymol	—	−5.7	−5.7	−5.6	−6.2	−6.3	−5.5	−4.7	−4.2
	Eugenol	—	−6.1	−5.6	−5.5	−5.5	−5.6	−6.0	−4.9	−3.7
	Caryophyllene	—	−7.5	−7.1	−6.5	−5.6	−7.9	−7.4	−5.5	−5.1
	Eugenol acetate	—	−6.5	−6.1	−6.2	−5.6	−6.1	−6.0	−4.7	−3.9
	cis‐Cinnamaldehyde	—	−5.4	−5.3	−5.1	−5.8	−5.7	−5.5	−4.4	−3.9
Standards	Ampicillin	—	−7.1	−8.0	−7.1	−6.5	−8.2	−6.4	−6.2	−5.0
	Erythromycin	—	−7.3	−8.4	−6.4	−5.8	−6.3	−5.5	−5.9	−4.4
	Ciprofloxacin	—	−8.4	−7.9	−7.5	−6.3	−7.3	−8.1	−6.3	−4.6

FIGURE 4.

2D interactions of caryophyllene with CmeB (A), CmeR (B), DHDPS (C), ErmB (D), GyrA (E), GyrB (F), PBP2 (G) and TetR (H) in C. jejuni.

FIGURE 5.

2D interactions of eugenol acetate with CmeB (A), CmeR (B), DHDPS (C), ErmB (D), GyrA (E), PBP2 (F) in C. jejuni.

In C. coli, caryophyllene exhibited favourable binding energies across most targets, with scores of −7.5, −7.1, −6.5, −7.9, −7.4, −5.5 and −5.1 kcal/mol for CmeB, CmeR, DHDPS, GyrA, GyrB, PBP2 and TetR, respectively (Table 7 and Figure 6). Specific interactions included π–π stacking and π–sigma bonds with residues such as ALA301 and PRO667 (CmeB), PHE137, VAL163 and TYR139 (CmeR), TYR110 (DHDPS), PRO787 and PHE777 (GyrA), VAL93 and ILE77 (GyrB), TYR467 (PBP2) and PHE6 (TetR). Towards ErmB, carvacrol exhibited a binding energy of −6.4 kcal/mol.

FIGURE 6.

2D interactions of caryophyllene with CmeB (A), CmeR (B), DHDPS (C), ErmB (D), GyrA (E), GyrB (F), PBP2 (G) and TetR (H) in C. coli.

It is important to emphasise that molecular docking predicts the thermodynamic likelihood of ligand–protein interactions but does not confirm biological inhibition or antimicrobial activity. The favourable binding energies observed for carvacrol, thymol, caryophyllene and eugenol acetate suggest potential interactions with several resistance‐associated targets. For instance, affinity towards CmeB may indicate possible interference with multidrug efflux, while interactions with GyrA and GyrB may relate to modulation of fluoroquinolone resistance mechanisms. Similarly, binding to PBP2 could suggest potential effects on peptidoglycan synthesis, and interaction with ErmB may imply engagement with the macrolide resistance pathway. Although certain phytochemicals displayed more favourable docking scores than the standard antibiotics against specific targets, binding energies of chemically unrelated molecules cannot be directly compared as evidence of biological superiority. These computational findings should be interpreted as hypothesis‐generating data that warrant further experimental validation through enzyme inhibition assays and mechanistic studies.

3.10. Pharmacokinetic Features

Molecular docking identified four compounds for ADME profiling: carvacrol, thymol, caryophyllene and eugenol acetate. Their predicted ADME and toxicity parameters are summarised in Table S2. All four compounds exhibited favourable aqueous solubility and high Caco‐2 permeability, with intestinal absorption rates exceeding 90% (caryophyllene: 94.85%; eugenol acetate: 94.76%; thymol and carvacrol: ∼90.84%). Skin permeability was predicted to be low for all compounds (log Kp > −2.5). None were predicted to be substrates or inhibitors of P‐gp, suggesting minimal transport‐related interference during absorption. Carvacrol, thymol and caryophyllene were predicted to exhibit high tissue distribution, whereas eugenol acetate showed the lowest Vd.

All compounds were predicted to cross the blood–brain barrier (logPS > −2). However, as these compounds are being investigated primarily as antimicrobials against poultry‐associated Campylobacter, central nervous system penetration is not a relevant or desired pharmacokinetic property in this context, and these predictions should be interpreted accordingly. None of the compounds were predicted to be substrates of CYP2D6 or CYP3A4. Thymol, carvacrol and eugenol acetate were predicted to inhibit CYP1A2, suggesting potential for selective metabolic interactions. All compounds were predicted to be non‐substrates of the renal OCT2 transporter and to have low renal clearance (log mL/min/kg < 3). Regarding toxicity predictions, none of the compounds were flagged for Ames mutagenicity or hERG I/II inhibition, indicating favourable cardiac safety profiles. Notably, thymol and carvacrol were predicted to exhibit hepatotoxicity, a finding consistent with reported hepatic effects of these monoterpenes in toxicological studies. All four compounds were predicted to have skin sensitisation potential.

4. Discussion

Our principal findings demonstrated a moderate prevalence of Campylobacter spp. in poultry in Morocco, a high prevalence of fluoroquinolone resistance, potent antibacterial activity of the tested EOs, and a synergistic interaction between OREO and ciprofloxacin.

In this study, the results indicated an occurrence of 15.52% of Campylobacter species. This occurrence is considered moderate, especially when compared to the higher rates reported by (Gharbi et al. 2018) in countries such as Spain (88%), Portugal (82%) and France (76.1%). However, it still exceeds the lower prevalence levels observed in certain Northern European countries, as also reported by the same study, including Sweden (13%), Finland (3.9%) and Denmark (10.3%). The moderate prevalence observed here may reflect factors limiting transmission within the flock, such as farm hygiene practices, biosecurity measures or antimicrobial use. In addition, several studies have shown that the risk of Campylobacter spreading in a flock is directly linked to the age of birds, and it increases with the presence of older flocks (Newell and Fearnley 2003). In our study, we only included birds aged 6–8 weeks, which may partly explain the low rate of Campylobacter spp.

Regarding the distribution of Campylobacter spp., in contrast to what is generally reported worldwide, C. coli was the most frequently isolated, accounting for 90% of the strains, while C. jejuni represented the remaining 10%. However, this finding aligns with the recent epidemiological situation observed in Morocco. Regional studies have reported a dominance of C. coli in poultry (Asmai et al. 2020; El Baaboua et al. 2021). This trend could be explained by C. coli's broader antimicrobial resistance profile, which gives it a survival advantage in farming environments. In terms of geographic distribution, no statistically significant difference was observed in Campylobacter positivity across the sampled regions, indicating the relatively uniform occurrence across the four surveyed regions.

A critical finding of our study is the high‐rate resistance of (quinolones/fluoroquinolones), namely, nalidixic acid (90%) and ciprofloxacin (85%), among Campylobacter isolates. Nevertheless, neither of these antibiotics is authorised for veterinary use in poultry in Morocco. The high resistance observed can be explained by the extensive use of enrofloxacin in veterinary medicine. This systematic overuse remains an important factor in Campylobacter resistance to fluoroquinolones. This drug, which is well absorbed and rapidly metabolised to ciprofloxacin in some animals, exerts a selection pressure that promotes the emergence of mutations within the gyrA gene, resulting in a rapid and concerning increase in ciprofloxacin‐resistant strains. The similar resistance rate between nalidixic acid and ciprofloxacin is highly expected due to cross‐resistance, as both drugs belong to the same antimicrobial family (fluoroquinolones/quinolones). At the genetic level, resistance to this class is caused mainly by point mutations occurring within the QRDR of the gyrase A (gyrA) gene, and partly by the multidrug efflux pump cmeABC, which extrudes the antibiotic and decreases its intracellular concentration (Mohan et al. 2025).

Tetracycline also reached an alarming rate of resistance (78%), likely because it is an authorised and widely used antibiotic in Morocco. A survey of antimicrobial consumption in broiler production in Morocco (Rahmatallah et al. 2018) revealed that enrofloxacin is the most commonly used antimicrobial, followed by tetracycline. Frequent exposure to these drugs is a key factor promoting the selection of resistant Campylobacter strains, a process that is further accelerated by the exchange of resistance‐associated genes among Campylobacter spp. This resistance is generally associated with the presence of tet(O), which was recognised as a primary mechanism of resistance because it encodes a protective ribosomal protein that facilitates the access of amino acids and transfer ribonucleic acids to the ribosome, an action that is normally blocked by tetracycline (Taylor et al. 1995).

In the case of erythromycin, the level of resistance reached 47%, indicating considerable antimicrobial pressure. Genetically, macrolide resistance is mostly linked to point mutations at positions 2074 and 2075 in the 23S rRNA, which alter the conformational structure of the binding pocket, thereby reducing the binding affinity of the antibiotic. (Luangtongkum et al. 2012). The overall resistance to streptomycin was moderate (36%), and no resistance was found towards gentamycin. However, both drugs are unauthorised for poultry use in Morocco. The lack of authorisation for use and the assumed low level of exposure to this antimicrobial may contribute to the observed low level of resistance in Moroccan Campylobacter spp.

Furthermore, GC–MS analysis revealed that OREO primarily consisted of thymol (77.8%) and carvacrol (19.62%), while CIEO contained mainly cis‐cinnamaldehyde (85.62%), and CLEO predominantly contained eugenol (82.11%). These composition profiles may explain the antimicrobial activity of the EOs tested, as well as the synergistic and antagonistic interactions observed with ciprofloxacin. Notably, the OREO–CIP combination demonstrated a synergistic effect. In conjunction with increased permeability and depolarisation of the cytoplasmic membrane (Xu et al. 2008), this synergistic effect can be attributed to the ability of thymol and carvacrol to inhibit efflux pumps, thereby preventing ciprofloxacin from being pumped out of the cytoplasm. This reasoning was demonstrated by Miladi et al. (2016), whose study highlighted the inhibition of efflux pumps by the same compounds and reported a synergistic effect with tetracycline against Salmonella enteritidis. More directly relevant, Oh and Jeon (2015) demonstrated that phenolic compounds reduced the expression of the multidrug efflux pumps CmeABC when combined with ciprofloxacin against C. jejuni, which supports the hypothesis of an efflux‐mediated mechanism for phenol‐fluoroquinolone synergy. However, this is not universal: de Sousa Silveira et al. (2020), in their study on Staphylococcus aureus, reported that carvacrol and thymol had no effect on efflux pumps and suggested that effects on other resistance mechanisms can be involved. Therefore, the underlying mode of action depends on the strain as well as the combination used rather than being an intrinsic property of the compound.

In contrast, the interaction between CLEO and ciprofloxacin was antagonistic. Given that the antimicrobial activity of this EO is attributed to its main compound. From a structural standpoint, eugenol is also considered a phenolic compound, like carvacrol and thymol; a similar synergistic effect with ciprofloxacin might have been expected, but our results did not confirm this. This discrepancy can be explained by the methoxy group present on eugenol, which limits its ability to liberate the phenolic proton and, consequently, reduces its membrane permeabilising capacity compared to carvacrol and thymol (Ben Arfa et al. 2006; Walczak et al. 2021). This structural limitation could explain why an antagonistic effect was observed with CLEO but not with OREO. Conversely, it has been reported that good synergistic effects exist between clove EO and ciprofloxacin against other species, such as Klebsiella pneumoniae and Acinetobacter baumannii (Jilani et al. 2026). However, Mukti et al. (2026) reported an antagonistic interaction between an ethanol‐based clove extract and ciprofloxacin against MDR E. coli strains. To our knowledge, no study has examined the combination of CLEO and ciprofloxacin specifically against Campylobacter species, which limits the ability to directly compare these findings and to elucidate the underlying mechanism.

Similarly, the interaction between CIEO and ciprofloxacin was antagonistic. Unlike OREO, CIEO is primarily composed of cis‐cinnamaldehyde, which belongs to the aldehyde family and therefore acts through a different mechanism by interfering with key enzymes involved in energy metabolism (Wang et al. 2021) explaining CIEO's intrinsic antibacterial activity, but not the antagonism observed in combination with ciprofloxacin. It should be noted that this antagonistic result contrasts with previous studies, which generally reported neutral interaction between cinnamon/cinnamaldehyde and ciprofloxacin, with no antagonism reported among P. aeroginosa isolates (Utchariyakiat et al. 2016). This divergence highlights that the antagonism observed with CIEO may be specific to the strains studied and EO composition, and that its underlying mechanism remains to be elucidated.

It is also worth noting that the composition of EOs varies by botanical species, geographical origin and batch‐to‐batch across the literature. Notably, carvacrol content was the dominant constituent in Moroccan Origanum compactum populations ranging from 0% to 96.3%, followed by thymol (0%–80.7%; Aboukhalid et al. 2016). Also, eugenol content in clove oil varies from 55.60% to 74.64% between Java and Manado in Indonesia (Amelia et al. 2017). Considering this variability in the content of major compounds, and given that only the composition of our EOs has been characterised, not their specific botanical species or origin, direct comparison with other studies regarding synergistic or antagonistic effects remains limited.

Furthermore, while EOs proved to have promising efficacy in vitro, the real challenge remains their practical application in poultry production. Their high volatility, low water solubility and poor intestinal absorption require nanoencapsulation strategies to improve their stability and bioavailability and to facilitate targeted delivery in poultry feed (Movahedi et al. 2024). Although these encapsulation strategies offer advantages, they incur high additional costs, which pose a significant economic obstacle. In addition to this economic barrier, it remains uncertain whether the MIC values obtained with our EOs are achievable at the target site once administered. The much‐documented problem associated with the use EOs as alternatives is that their active constituents often reach the intestines at concentrations below the inhibitory threshold required for antimicrobial activity (Qui 2023), although this specific concentration reached in vivo is rarely quantified. One of the few studies to provide such information reported intestinal digesta concentrations of only 0.20–0.80 µg/g (wet weight) for thymol and carvacrol following dietary supplementation of 60–240 mg/kg feed in broilers. This concentration was not enough to reduce C. perfringens numbers in vivo, contrary to the strong activity against this pathogen in vitro (Du et al. 2015). This discrepancy illustrates the gap that must be addressed before our EO–antibiotic combinations' practical application in poultry production. Regarding safety, dietary EO supplementation at doses of 60–240 mg/kg was well tolerated in broilers, with no adverse effects on growth performance (Du et al. 2015), implying that safety is probably not the limiting factor for practical application. However, regulatory oversight of OEs remains limited, although most of the EOs are generally recognised as safe (GRAS) by the FDA (U.S. FDA 2026); no US regulatory agency provides certification or approval for EOs in terms of their quality or purity (Fontana et al. 2025). Within the European Union, the EFSA's FEEDAP Panel was unable to complete a full risk assessment of food additive derived from oregano EO, due to incomplete compositional characterisation (EFSA Panel on Additives and Products or Substances used in Animal Feed (EFSA FEEDAP Panel) 2019).

These findings underscore the importance of in vivo pharmacokinetic and toxicity studies, which are therefore warranted before EO–antibiotic combinations can be considered for practical application in poultry production.

Among the docked phytochemicals, caryophyllene consistently exhibited the most favourable binding energies across the majority of resistance‐associated targets in both C. jejuni and C. coli, particularly towards CmeB (−7.5 kcal/mol), CmeR (−7.5 kcal/mol), GyrB (−7.4 kcal/mol) and GyrA (−7.9 kcal/mol in C. coli). Thymol showed the strongest affinity for RNApα (−6.7 kcal/mol), an essential subunit crucial for transcriptional regulation (Zakharova et al. 1998).

Carvacrol demonstrated selective binding towards ErmB in both species (−6.1 and −6.4 kcal/mol). Macrolide resistance has been linked to nucleotide mutations at positions 2058 and 2059 in the 23S rRNA peptidyl transferase region (Niwa et al. 2001; Vacher et al. 2003), and engagement with ErmB may imply interference with this resistance pathway. Eugenol acetate displayed notable affinity for PBP2 (−6.6 kcal/mol), a key enzyme in peptidoglycan cross‐linking that contributes to bacterial cell wall integrity (Choi et al. 2024), forming conventional hydrogen bonds with HIS554 and GLU421. These differential binding profiles suggest that the antibacterial activity observed in our experimental assays likely stems from a multi‐target interaction strategy rather than single‐target inhibition.

The strongest synergism observed experimentally with OREO compared to the other EOs may be partially explained by the docking performance of its major constituent, carvacrol. Although caryophyllene exhibited the broadest target coverage, carvacrol's strong experimental efficacy combined with its favourable binding to ErmB and moderate affinities across efflux and cell wall targets suggests a complementary mechanism. Conversely, cinnamaldehyde showed comparatively weaker affinities despite its strong antimicrobial activity assessed in vitro. This discrepancy suggests that CIOE's efficacy relies on alternative mechanisms of action that are not captured by the selected in silico approach. The presence of caryophyllene as a shared major component across oregano and clove EOs likely contributes to the overall observed efficacy through its broad‐spectrum docking profile. From a mechanistic standpoint, the most promising targets appear to be CmeB and CmeR, components of the CmeABC multidrug efflux system, which contributes significantly to broad‐spectrum antibiotic resistance (Lin et al. 2002, 2005). Caryophyllene's strong affinity towards both proteins (−7.5 kcal/mol) suggests a potential to interfere with efflux pump function or its transcriptional regulation, thereby potentially restoring antibiotic susceptibility by increasing intracellular drug accumulation. Additionally, DNA gyrase, composed of subunits encoded by gyrA and gyrB, is essential for DNA supercoiling and represents the primary target of fluoroquinolones (Champoux 2001; Han et al. 2012). Favourable binding to GyrA and GyrB may relate to modulation of fluoroquinolone resistance mechanisms, particularly given that resistance primarily results from mutations in gyrA, notably the Thr‐86‐Ile substitution (B. Ge et al. 2005). Interaction with PBP2 could indicate potential effects on peptidoglycan cross‐linking and cell wall integrity (Choi et al. 2024), while the enzyme DHPS, essential in folate biosynthesis (Gibreel and Sköld 1999), also represents a validated antimicrobial target. Caryophyllene's favourable binding to TetR (−7.1 kcal/mol), a regulator of the CmeABC efflux system (Ramos et al. 2005), suggests potential dual interference with both pump expression and transcriptional control. However, these interpretations remain speculative; molecular docking predicts binding likelihood but cannot confirm functional inhibition, and the thermodynamic favourability of a ligand–protein complex does not necessarily translate into biological activity. The predicted pharmacokinetic profiles of carvacrol, thymol, caryophyllene and eugenol acetate revealed high intestinal absorption (> 90%) and favourable tissue distribution for three of the four compounds. By analysing molecular structure and physicochemical properties, one can predict how a candidate compound behaves in biological systems, including absorption, distribution, metabolism and excretion (Saidi et al. 2022). The role of P‐gp, a membrane transporter responsible for pumping substances out of cells, was also considered to evaluate its possible influence on these compounds (Constantinides and Wasan 2007). The absence of P‐gp substrate status and the low predicted renal clearance further support efficient absorption and retention. Three compounds exhibited high Vds, suggesting that these molecules can effectively reach and distribute within target tissues (Pires et al. 2015), although eugenol acetate showed the lowest value. Such favourable pharmacokinetic profiles may highlight their potential for further therapeutic development (Varma et al. 2015). Notably, all compounds were predicted to cross the blood–brain barrier; however, given that these EOs are being investigated primarily as antimicrobials against poultry‐associated Campylobacter, central nervous system penetration is neither a relevant nor a desired property in this context. Regarding toxicity predictions, the Ames test, a standard bacterial assay used to assess the mutagenic potential of chemical compounds, was considered to evaluate their toxicity (Turkez et al. 2017). None of the compounds were flagged for Ames mutagenicity or hERG I/II inhibition, indicating favourable cardiac safety profiles. Notably, thymol and carvacrol were predicted to exhibit hepatotoxicity. This predicted liability aligns with experimental evidence from rodent toxicology studies: thymol has been shown to induce dose‐dependent hepatic enzyme elevations and hepatocellular vacuolisation in rats at concentrations exceeding 100 mg/kg/day (Gaudio et al. 2025), while carvacrol administration has been associated with mild hepatic injury and altered serum transaminase levels in sub‐chronic exposure models (Ibrahim et al. 2026). These findings suggest that the hepatotoxicity prediction is not merely an in silico artefact but reflects a documented biological liability of phenolic monoterpenes. The skin sensitisation potential predicted for all four compounds also warrants attention if topical or handling exposure is anticipated in farm settings (Suntres et al. 2015; Tomsuk et al. 2024). Several limitations should be acknowledged. First, molecular docking provides a static prediction of ligand–protein interactions and does not account for protein conformational flexibility, solvent effects or cellular permeability; therefore, the binding energies reported here cannot be equated with biological inhibition. Second, the exclusion of trace constituents (< 1% abundance) from the docking analysis may overlook synergistic or additive interactions that contribute to the overall antibacterial effect of the EOs. Third, the use of an E. coli GyrA surrogate for C. jejuni, while justified by QRDR sequence conservation, may not fully capture species‐specific allosteric differences. Fourth, Lipinski's rule of five has limited applicability to volatile, lipophilic phytochemicals, and its use here was intended as a preliminary screening tool rather than a definitive predictor of antimicrobial potential. Finally, the docking protocol, while validated by re‐docking where possible, would benefit from additional experimental confirmation through enzyme inhibition assays, surface plasmon resonance or site‐directed mutagenesis studies. These computational findings should therefore be interpreted as hypothesis‐generating data that guide, but do not replace, experimental mechanistic validation.

4.1. Limitations of the Study

This study has several limitations that should be acknowledged. First, only ciprofloxacin was evaluated in combination with EOs, so the finding cannot be generalised to other antibiotic classes without further testing. Second, Agar diffusion assays were used for preliminary screening of antibacterial activity so the MIC and MBC values reported here provide the quantitative basis for our conclusions. Third, species identification and the distinction between C. jejuni and C. coli were performed using MALDI‐TOF mass spectrometry. Although this method is highly reliable, recognised by ISO standards, and systematically validated in our laboratory using rigorous identification scores, the lack of molecular confirmation (such as PCR targeting species‐specific genes like hipO or ceuE) may be considered a minor limitation. Fourth, resistance mechanisms in the tested strains proposed in the discussion remain hypothetical and should be validated through targeted molecular assays in future work. Finally, it is currently difficult to correlate our in vitro MIC values (% v/v) with achievable concentrations in the poultry gastrointestinal tract. This requires in vivo studies assessing pharmacokinetic safety and efficacy.

5. Conclusion

This study provided essential data on antibiotic resistance in strains of avian Campylobacter spp. isolated from selected regions of Morocco. The findings reveal concerning resistance rates within these poultry isolates, particularly highlighting a high prevalence of quinolone/fluoroquinolone resistance, namely, NAL (90%) and CIP (85%), along with TET (78%), indicating a potential food safety concern associated with poultry consumption in these regions. These data underline the importance of ongoing surveillance and the implementation of rational antibiotic management strategies. Meanwhile, in vitro testing of the antibacterial activity of CZEO, SAEO and OCEO showed promising results. The commercial EOs studied demonstrated significant antibacterial activity against Campylobacter spp. Furthermore, the combination of OCEO with CIP demonstrated a synergistic effect, indicating a potential to enhance the efficacy of this antibiotic. The molecular docking revealed strong binding affinities between the major EO components, namely, caryophyllene as well as eugenol acetate, and key resistance‐related targets, predicting possible interactions. Finally, future studies should focus on in vivo models and experimental validation to fully elucidate these mechanisms of action and evaluate the feasibility of their application in practical poultry production.

Author Contributions

Zineb Soubai: investigation, writing – original draft, data curation, methodology, conceptualisation. Nadia Ziyate: methodology, investigation, writing – review and editing, data curation. Rim Rais: methodology, data curation, investigation. Hamza Elhrech: data curation, methodology, investigation. Oumayma Aguerd: methodology, data curation, investigation. Abdelhakim Bouyahya: formal analysis, writing – review and editing, validation, methodology, investigation. Taoufiq Benali: methodology, data curation. Mohamed Akhazzane: methodology, data curation. Hiba Mahir: writing – review and editing, methodology, software. Siham Fellahi: methodology, writing – review and editing, investigation. Waleed Al Abdulmonem: funding acquisition, investigation, visualisation, writing – review and editing, validation, methodology. Imane El Houssni: formal analysis, writing – review and editing, validation, investigation. Benaissa Attarassi: supervision, writing – review and editing, conceptualisation, methodology. Nabila Auajjar: conceptualisation, writing – review and editing, methodology, project administration, supervision.

Funding

The researchers would like to thank the Deanship of Graduate Studies and Scientific Research at Qassim University (www.qu.edu.sa) for financial support (QU‐APC‐2026). This study was also supported by the IAEA (Research Project CRP D52044), which provided all materials for bacterial work.

Ethics Statement

The study involved the use of post mortem samples from animals destined for consumption, following Moroccan food safety law 28‐07. Since there was no live animal involvement, IACUC approval was not required. Collection and handling were carried out by standard ISO 10272‐1:2017 to avoid cross‐contamination. The approach aligns with the 3Rs principle of ethically using by‐products from the food industry.

Consent

The authors have nothing to report.

Supporting information

Supporting File: vms371256‐sup‐0001‐SuppMat.DOC

Data Availability Statement

The data that supports the findings of this study are available in the supplementary material of this article.