Antimicrobial Resistance and Species Dynamics of Enterococcus faecalis and Enterococcus faecium in Breeding Hens: Phenotypic and Genotypic Characterization

1. Introduction

Enterococcus spp. are Gram-positive bacteria widely distributed in the gastrointestinal microbiota of humans and animals, as well as in environmental reservoirs. Due to their persistence and fecal origin, they are commonly used as indicators of contamination and antimicrobial resistance (AMR) surveillance [1]. At the same time, enterococci are opportunistic pathogens of increasing relevance, particularly E. faecalis and E. faecium, which are the most frequently isolated species from animal samples [2], frequently associated with severe infections in humans, including septicaemia, endocarditis, and urinary tract infections [3].

In poultry production systems, enterococci are part of the normal intestinal microbiota but can also act as opportunistic pathogens, contributing to conditions such as septicaemia or skeletal disorders, with potential economic impact on flock productivity [4]. In addition to their veterinary relevance, Enterococcus spp. are considered key indicator bacteria for monitoring AMR in food-producing animals [1] because they exhibit intrinsic resistance to several antibiotics and are able to acquire additional resistance through plasmid and transposon transfer, as well as chromosomal mutations [3]. Antibiotic-resistant enterococci have been isolated worldwide from both poultry and poultry products [5,6].

The poultry industry represents one of the largest and fastest-growing sectors of animal protein production worldwide [7]. The use of antimicrobials for therapeutic, prophylactic, and sometimes production-related purposes has contributed to the emergence and dissemination of antimicrobial resistance in poultry-associated bacteria. In particular, tetracyclines and macrolides have historically been widely used, and resistance to these antimicrobial classes remains highly prevalent in enterococci isolated from poultry systems [3,5,8].

From a veterinary perspective, the presence of antimicrobial-resistant enterococci in poultry production is particularly relevant as these bacteria can colonize animals throughout their life cycle, persist in the farm environment, and spread within and between flocks. Moreover, management practices, housing systems, and feeding strategies can influence both gut microbiota composition and the selection of resistant strains [9].

From a One Health vision, antimicrobial resistance in enterococci from poultry is a question of veterinary and public health concern, as resistant bacteria and resistance genes may be transmitted to humans through the food chain, direct contact, or the environment [1,10]. Recent reviews emphasize that poultry production is a key interface where antimicrobial resistance can emerge and disseminate, requiring integrated surveillance strategies across sectors [11].

Despite the large body of research on antimicrobial resistance in broilers, there is comparatively limited information on breeding and laying hens, especially under commercial production conditions [12,13]. This lack of data represents a significant knowledge gap, as breeding and laying hens occupy a critical position in the poultry production pyramid and may act as reservoirs of antimicrobial resistance that can be transmitted vertically or horizontally across the production chain [14,15].

Therefore, the aim of this study was to investigate the prevalence of antimicrobial resistance and the distribution of macrolide and tetracycline resistance genes in Enterococcus faecalis and Enterococcus faecium isolated from breeding hens in commercial poultry farms in Spain, contributing to the understanding of antimicrobial resistance within the poultry production pyramid and its implications under a One Health framework.

2. Materials and Methods

2.1. Sampling

Two breeding hen farms located in eastern Spain were included in the study. In farm A, a flock of 14,500 hens (Gallus gallus) was sampled at five time points: upon arrival at the farm (SA1; one-day-old chicks), during the rearing phase (SA2 and SA3; 4- and 19-week-old pullets), and during the production phase (SA4 and SA5; 25- and 28-week-old hens), prior to their transfer to the production farm. During the study period, tylosin and amoxicillin were administered for therapeutic purposes at week 4 (5-day treatment) and week 25 (4-day treatment), respectively.

In farm B, the batch consisted of 4,100 adult breeder hens and 408 males. Samples were taken ten times at three-week intervals (SB1–SB10). Birds were 27 weeks old at the first sampling and 54 weeks old at the last one. A treatment with enrofloxacin during 5 days was administered by authorized veterinarians at SB4 sampling point (36-week-old animals).

The samples were collected in accordance with the guidelines of the “National Programme for the surveillance and control of certain Salmonella serotypes in flocks of breeding hens of the species Gallus gallus” [16], to ensure that the samples were representative of the entire flock.

Sampling SA1 consisted of four samples of 25 g each collected from transport box containing meconium from one-day-old chicks. All other samplings in both farms consisted of a pair of absorbent boot swabs impregnated with the soil residues of the farm and two composite faecal samples (approximately 100 g of manure collected from belt systems). All samples were refrigerated and processed within 24 hours.

Samples were diluted in buffered peptone water (Buffered Peptone Water, ISO, Scharlau, Spain). In SA1, 25 g of sample were mixed with 225 mL of peptone water. Boot swabs were weighed and mixed in 1:10 w/v peptone water. For composite faecal samples, 25 g were taken, diluted and homogenized in 225 mL of peptone water using a stomacher (BagMixer, Interscience, France).

2.2. Isolation and Identification of Enterococcus spp.

After homogenization, serial decimal dilutions were prepared and plated on Slanetz–Bartley agar (Scharlab S.L., Spain) by inoculating 0.1 mL. Plates were incubated at 37 °C for 48 h. Colonies displaying typical Enterococcus colour and morphology were randomly selected and streaked onto Brain Heart Infusion Agar (BHIA).

Thirty colonies were selected at SA1, whereas up to 15 colonies per sample type (faeces and boot swabs) were selected for the remaining sampling points across both farms. After incubation for 24 h at 37 °C, isolates were further confirmed by incubation for 2 h at 42 °C on kanamycin–aesculin–azide agar (KAA agar), selecting colonies showing a black halo surrounding growth.

Confirmed Enterococcus spp. isolates were identified as E. faecalis or E. faecium by multiplex PCR targeting the ddl gene, which encodes D-Ala–D-Ala ligase, using primers and conditions described by Depardieu et al. [17] (Table 1).

Briefly, the mPCR assays were carried out in a final volume of 25 µL with the following reagent concentrations: 1× NH₄ Reaction Buffer (BIOTAQ™ DNA Polymerase, Bioline), 0.5 mM of each dNTP (DNTP Mix 100 mM, Bioline), 1.5 mM MgCl₂ (BIOTAQ™ DNA Polymerase, Bioline), 1.25 U Taq DNA polymerase (BIOTAQ™ DNA Polymerase, Bioline), and the primer concentrations as specified by the authors. Amplification was carried out as follows: 3 min at 94 °C and 30 cycles of amplification consisting of 1 min at 94 °C, 1 min at 54 °C, and 1 min at 72 °C, with 7 min at 72 °C for the final extension.

Confirmed isolates were stored at -21 °C until further analysis.

2.3. Antimicrobial Susceptibility Testing

Antimicrobial susceptibility was determined using the disk diffusion method on Mueller-Hinton agar (Scharlau S.L., Spain) following CLSI [18] guidelines. Plates were incubated at 37 °C for 18 hours.

The antibiotics tested were selected according to EFSA recommendations [19]: erythromycin (15 µg), vancomycin (30 µg), gentamicin (120 µg), streptomycin (300 µg), quinupristin/dalfopristin (15 µg), ampicillin (10 µg), tetracycline (30 µg), chloramphenicol (30 µg), and ciprofloxacin (5 µg). High-content aminoglycoside disks were used to detect high-level resistance, since enterococci exhibit intrinsic low-level resistance and may appear susceptible in vitro while requiring higher therapeutic doses in vivo [20]. E. faecalis was not tested for quinupristin/dalfopristin due to intrinsic resistance [2].

Resistance levels were classified according to EFSA and ECDC [1] as follows: rare (<0.1%), very low (0.1–1.0%), low (>1–10%), moderate (>10–20%), high (>20–50%), very high (>50–70%), and extremely high (>70%).

2.4. Molecular Analysis

DNA extraction was performed using the GenElute Bacterial Genomic DNA Kit (Sigma-Aldrich, USA). DNA samples were stored at −21 °C.

Resistance genes to tetracyclines and macrolide–lincosamide–streptogramins (MLS) were detected by PCR. The genes analysed were ermA and ermB for MLS resistance, and tetK, tetL, tetM, tetO, and tetS for tetracycline resistance.

PCR conditions for MLS genes were those described by Chen et al. [21], while tetracycline gene amplification followed Ng et al. [22] protocol. Reactions were performed in a final volume of 25 µL containing 1× NH₄ reaction buffer, 0.5 mM dNTPs, MgCl₂ (2.5 mM for MLS and 3 mM for tetracycline genes), 1.25 U Taq DNA polymerase, and primers at the concentrations specified in Table 2.

2.5. Statistical Analysis

Statistical analyses were made using “Statgraphics Centurion XVII” (Statpoint Technologies Inc., Warrenton, Virginia) software. Categorical variables (species distribution, antimicrobial resistance, resistance profiles, and gene prevalence) were compared using χ² test or Fisher’s exact test when expected frequencies were low. Non-parametric Mann–Whitney U test was used for comparisons of the number of resistances per isolate. To account for multiple comparisons, Bonferroni correction was applied. Statistical significance was set at p < 0.05.

3. Results

3.1. Isolation of Enterococcus spp.

A total of 150 isolates were obtained from farm A. PCR identification showed that 103 isolates (68.7%) were Enterococcus faecium and 47 (31.3%) were Enterococcus faecalis. When analysed by age group, the distribution of species varied significantly across production stages (χ² test, p < 0.001): All isolates obtained from one-day-old chicks (30 isolates) were identified as E. faecalis. Among pullets (60 isolates), 80% (48) were identified as E. faecium and 20% (12) as E. faecalis. In adult hens (60 isolates), 91.7% (55) corresponded to E. faecium and 8.3% (5) to E. faecalis.

In farm B (adult hens) a total of 180 isolates were obtained, of which 160 (88.9%) were identified as E. faecium and 20 (11.1%) as E. faecalis.

A significant difference in species distribution was observed between farms (χ² test, p = 0.00001), with a higher odds of isolating E. faecalis in farm A (OR = 3.65, 95% CI: 2.08–6.40).

3.2. Antimicrobial Resistance

The full distribution of resistance percentages is shown in Table 3. Overall, isolates from both farms showed resistance mainly to tetracycline and, to a lesser extent, erythromycin.

In E. faecalis isolates from farm A, extremely high resistance levels were detected for tetracycline and very high levels for erythromycin, ciprofloxacin and chloramphenicol. In E. faecalis isolates from farm B, very high resistance to tetracycline and low resistance to erythromycin were detected, while resistance to other antibiotics was either absent or low. No isolate from both farms was resistant to ampicillin, vancomycin, gentamicin or streptomycin.

Significant differences between farms were observed for erythromycin resistance, which was markedly higher in farm A compared to farm B (p < 0.001, 95% CI: 30–69%), and for ciprofloxacin and chloramphenicol resistance, which were present only in farm A. No significant differences were found for tetracycline resistance.

E. faecium isolates from farm A showed extremely high resistance to tetracycline, very high resistance to erythromycin, low resistance to quinupristin/dalfopristin and ciprofloxacin, and very low resistance to ampicillin. In farm B, extremely high resistance to tetracycline, moderate resistance to erythromycin, and very low resistance to ciprofloxacin and quinupristin/dalfopristin were observed. Erythromycin resistance was significantly higher in farm A than in farm B (p < 0.001, 95% CI: 30–52%).

When both species were considered jointly, resistance prevalence differed significantly between farms (χ² test, p < 0.001; OR ≈ 2.1, 95% CI: 1.6–2.8), indicating a higher overall burden of antimicrobial resistance in farm A.

3.3. Antimicrobial Resistance Profiles of Isolates

Resistance profiles were determined for all isolates exhibiting at least one resistance (Table 4).

Among E. faecalis isolates from farm A, 59.6% were resistant to four antibiotics, whereas in farm B, most isolates were resistant to either one antibiotic or none. For E. faecium, most isolates in farm A were resistant to one or two antibiotics, while in farm B, the majority showed resistance to only one antibiotic. In both farms, resistance to more than three antibiotics was rare. The distribution of resistance counts per isolate differed significantly between farms for both species, with significantly higher values observed in farm A (p < 0.001).

Tetracycline resistance was the most common trait in all the profiles, frequently detected either alone or in combination with erythromycin resistance. Only two resistance profiles were identified among E. faecalis isolates in both farms. In farm A, the most common profile was E-TE-C-CIP, whereas in farm B the predominant profile was TE alone. Among E. faecium isolates, six profiles were observed in farm A, with E–TE and TE being the most frequent, while in farm B, five profiles were identified, with TE alone being the most frequent pattern.

Multidrug resistance (MDR), defined as resistance to three or more antimicrobial classes [23] was detected in 59.6% of E. faecalis isolates in farm A, whereas no MDR isolates were found in farm B. In E. faecium, MDR prevalence was lower (12.6% in farm A and 1.3% in farm B).

3.4. Antibiotic Resistance Genes

The prevalence of resistance genes is shown in Table 5. Among MLS genes, ermB was the most prevalent in both farms, being detected in all E. faecalis isolates and in 94.8% and 34.6% of E. faecium isolates in farms A and B, respectively. The ermA gene was only detected in E. faecalis from farm A.

For tetracycline resistance, tetK, tetL, and tetM were the most prevalent genes. tetL and tetM were present in all E. faecalis isolates from farm A, while in farm B, tetK, tetL, tetM, and tetO were also detected. In E. faecium isolates, tetK, tetL, and tetM were the most frequently detected genes in both farms.

Taken together, gene distribution differed significantly between farms (χ² test, p < 0.001), mainly due to differences in the prevalence of ermA (36.0% in farm A vs. 0% in farm B), ermB (markedly more frequent in farm A, p < 0.001; OR = 42.76, 95% CI: 10.1–181.2), and tetK (significantly more prevalent in farm B, p = 0.041; OR = 0.33, 95% CI: 0.12–0.92).

When data from both farms were pooled, comparisons revealed significant differences between species for ermB (Fisher’s exact test, p = 0.0016; OR ≈ 3.5) and tetK (p < 0.001; OR << 1), both associated with E. faecium.

Multiple resistance genes commonly co-occurred within the same isolates, with profiles including ermB–tetL–tetM being the most predominant (Table 6). Both in farm A and B, E. faecalis isolates carrying more than one gene showed a reduced diversity of profiles, with nearly all strains carrying the combination ermA–ermB–tetL–tetM (farm A) and ermB-tetK-tetL-tetM (farm B). E. faecium displayed a greater diversity of profiles, with a lower predominance of ermB-associated profiles and a higher proportion of tetracycline-only gene combinations in farm B.

Significant differences were also observed in the prevalence of ermB among E. faecium isolates from farm A compared to farm B (p < 0.001; OR = 4.5, 95% CI: 2–9), being significantly lower in the latter. For E. faecalis, no meaningful statistical comparisons could be performed due to the very low number of isolates from farm B.

4. Discussion

This study provides new insights into antimicrobial resistance patterns in Enterococcus faecalis and Enterococcus faecium from breeding hens, a little studied stage in the poultry production pyramid. A key finding is the age-related shift from E. faecalis in early stages to E. faecium in adult hens. This ecological shift was statistically supported, with significant differences both between age groups within farm A and between farms, suggesting that host age and production stage play a major role in shaping enterococcal populations. However, as we did not perform a longitudinal study tracking the same animals throughout the entire production cycle in both farms, our findings should be interpreted as temporal changes across production stages rather than true longitudinal dynamics at the individual level.

Experimental studies have shown that antimicrobial exposure, including tylosin, may promote a shift in enterococcal populations towards E. faecium, likely due to its higher intrinsic and acquired resistance capacity [24]. Moreover, recent studies indicate that E. faecium has greater ecological adaptability in intensive systems, including stress tolerance and an enhanced capacity to acquire resistance determinants [25]. Its predominance in adult hens may therefore reflect both resistance and ecological fitness advantages, as also observed in European systems [5,8]. This shift has important One Health implications, since E. faecium is a major reservoir of antimicrobial resistance genes and is frequently associated with nosocomial infections [26]. Its dominance in adult birds increases the epidemiological relevance of poultry as a resistance reservoir.

High resistance rates were observed for tetracyclines, in agreement with [27], reinforcing its role as a key driver of resistance selection in poultry. Despite reduced antimicrobial use in Europe, surveillance studies indicate that tetracycline resistance remains among the highest [28]. Similar patterns have been reported in other Spanish poultry systems [5].

Although significant differences in resistance prevalence were observed between farms, these differences are likely influenced by variations in age distribution and production stage rather than reflecting intrinsic farm-specific effects.

Comparison between farms also revealed distinct gene distribution patterns. Farm A showed a higher prevalence and diversity of resistance genes, particularly among E. faecalis, in which ermA, ermB, tetL, and tetM were detected in 100% of isolates. In contrast, farm B exhibited a reduced gene spectrum, especially for E. faecium, suggesting differences in selective pressure between production environments.

In contrast, resistance to clinically relevant antibiotics such as vancomycin, high-level aminoglycosides, streptomycin and ampicillin was rare or absent, consistent with previous reports indicating lower resistance in animal isolates compared to human ones [29]. In Spain, resistance to vancomycin and chloramphenicol in E. faecium has been reported as rare, although higher resistance to quinupristin/dalfopristin has been described elsewhere, differing from our results [5,12]. For E. faecalis, moderate or high resistance to streptomycin has also been previously reported [30,31] while in our study, no resistance was detected. The absence of these resistances is consistent with the limited use of these antimicrobials in poultry production and suggests a relatively low selective pressure for these antibiotics in this context.

Despite the administration of some short-term antimicrobial treatments such as tylosin and amoxicillin, no noticeable increase in resistance levels was observed after them. This suggests that short-term therapeutic exposure may not be sufficient to significantly alter resistance patterns in enterococci populations, as suggested by other authors [32,33].

Multidrug resistance (MDR) levels were lower than in previous studies [30,34]. This limited accumulation of resistance traits also suggests relatively limited selective pressure. When present, MDR almost always involved tetracycline, often combined with erythromycin. The frequent co-occurrence of tetracycline and erythromycin resistance suggests co-selection mechanisms, supported by the detection of ermB, tetL, and tetM genes. Multiple resistance genes were frequently detected within single isolates, even with limited MDR, also supporting the role of co-selection [35].

The predominance of ermB, tetL, and tetM aligns with genomic studies identifying these as widespread resistance determinants and describing their association with mobile elements such as plasmids and transposons [8], thus facilitating horizontal transfer and persistence even without antimicrobial pressure and increasing dissemination potential along the food chain.

The frequent co-detection of tetL and tetM is particularly noteworthy, as it indicates the coexistence of different resistance mechanisms (efflux pumps and ribosomal protection), potentially increasing resistance stability under varying environmental conditions. Taken together, these findings highlight the need to monitor not only phenotypic resistance but also the underlying genetic determinants [36].

Overall, farm-related factors significantly influenced both the distribution of Enterococcus species and the resistance burden. Isolates from farm A, which included younger birds, exhibited a significantly higher resistance levels, both in terms of prevalence and number of resistances per isolate. This pattern was confirmed by the analysis of resistance counts per isolate, which showed significantly higher values in farm A for both species. In addition, several antibiotics, including ciprofloxacin and chloramphenicol were exclusively detected in farm A. Those differences between farms were statistically significant. However, these variations are likely influenced by differences in bird age and production stages rather than consistent farm-specific factors. Therefore, caution is required when attributing these differences to management or environmental conditions alone.

Multiple resistance genes frequently co-occurred within the same isolates, particularly combinations involving ermB, tetL, and tetM, highlighting the widespread presence of stable resistance gene clusters.

Interestingly, MDR prevalence was higher in E. faecalis than in E. faecium in farm A, contrasting with previous reports where E. faecium is often the primary MDR species [37,38].

Taken together, these findings support the hypothesis that younger birds may initially carry a higher load of antimicrobial resistance determinants, which may decrease over time in the absence of sustained selective pressure. This interpretation is consistent with previous studies showing that newly hatched chicks can harbour bacteria with high levels of antimicrobial resistance. E. coli isolated from the meconium of one-day-old chicks has been shown to exhibit high resistance rates [14,39], supporting the idea that antimicrobial resistance may be introduced early in the production chain. Such early acquisition of resistance may occur through vertical transmission from parent flocks or through contamination during hatching. Proposed mechanisms include infection of the reproductive tract during egg formation or faecal contamination at oviposition, as well as transmission at hatchery level [14,40,41]. Under this framework, the higher resistance levels observed in farm A may reflect the persistence of these early-acquired resistance traits, which could progressively decline as birds age and are no longer exposed to the same selective pressures.

Our findings strongly suggest that the antimicrobial resistance burden in breeding hens’ farms is not only related to immediate antimicrobial use but also by the temporal dynamics of resistance acquisition and loss throughout the production cycle. Variations in antimicrobial practices, biosecurity measures, or environmental conditions may further contribute to these patterns, although such factors were not directly assessed in this study. Overall, the results highlight the importance of considering bird age and antibiotic exposure together with environmental or production factors when interpreting antimicrobial resistance patterns in poultry production systems.

Several limitations should be considered when interpreting our findings. Sample size imbalance between farms, particularly for E. faecalis in farm B, may have reduced statistical power. Additionally, although multiple sampling points were included, isolates originated from the same flocks, which may introduce potential clustering effects, thus limiting statistical independence.

Future research should focus on genomic characterization of resistance determinants, the role of mobile genetic elements in their dissemination, and the influence of specific farm practices on the selection and persistence of antimicrobial resistance.

5. Conclusions

This study demonstrates that antimicrobial resistance in E. faecalis and E. faecium from breeding hens is generally low at the phenotypic level, except for tetracycline and erythromycin. However, the widespread presence and co-occurrence of resistance genes highlight the potential for persistence and dissemination of antimicrobial resistance. The observed species shift across production stages, and the stability of resistance patterns reinforce the importance of including breeding hens in AMR surveillance strategies. These findings support a One Health approach, emphasizing the need to integrate phenotypic and genotypic data to better understand and control antimicrobial resistance within poultry production systems.