Skip to main content
NCBI home page
Environmental Microbiology Reports logoLink to Environmental Microbiology Reports
. 2026 Sep 16;18(5):e70416. doi: 10.1111/1758-2229.70416

Antimicrobial Resistance in Nontyphoidal Salmonella and Clinically Relevant Enterococcus From Faecal Samples of Conservation‐Priority Captive Ungulates in a United Arab Emirates Urban Zoo: A Cross‐Sectional Baseline Study

Ihab Habib 1,2,✉, Moneeb A Qablan 1,2, Fatema Aaref Alzaabi 1, Fatema Rashed Alnuaimi 1, Ghalya Ahmed Almemari 1, Glindya Bhagya Lakshmi 1,2, Safeya Al Naqbi 1,2, Fatima Al Shamsi 1,2, Dhanya Vijay 1,2, John Mhongovoyo 3, Sultan Khalifa Alkaabi 3, Amna Mana Alotaiba 3
PMCID: PMC13583175  PMID: 42750254

ABSTRACT

Antimicrobial resistance (AMR) is a One Health challenge driven by microbial exchange among humans, animals and the environment. Zoological institutions offer useful settings for environmental AMR surveillance. This single‐zoo cross‐sectional study examined the occurrence, antimicrobial susceptibility and genomic characteristics of nontyphoidal Salmonella enterica (NTS) and clinically relevant Enterococcus spp. in faecal samples from 101 clinically healthy captive ungulates representing seven conservation‐priority species at a major urban zoo in the United Arab Emirates. NTS was detected in 4/101 samples (3.9%), including serovars Schwarzengrund (n = 2), Kentucky (n = 1) and Chester (n = 1). Among the four recovered NTS isolates, all met the study MDR definition within the tested panel, including a Salmonella Kentucky ST198 isolate carrying multiple resistance genes and quinolone‐associated mutations. Enterococcus spp. were detected in 77/101 samples (76.2%), dominated by Enterococcus faecium and Enterococcus casseliflavus (each 41.5%). Among 33 E. faecium/Enterococcus faecalis isolates tested phenotypically, resistance was generally low, with erythromycin and ciprofloxacin resistance each observed in 9.1%. One clinically important E. faecium isolate showed glycopeptide resistance and genetic markers associated with reduced daptomycin susceptibility. These single‐institution cross‐sectional data provide an initial regional baseline for AMR‐relevant enteric bacteria in conservation‐managed ungulates and identify priorities for broader longitudinal and interface‐based surveillance.

Keywords: antimicrobial resistance, environmental surveillance, one health, Salmonella enterica , zoological institutions

Targeted faecal surveillance of conservation‐priority captive ungulates identified multidrug‐resistant nontyphoidal Salmonella and clinically relevant Enterococcus in a UAE urban zoo. These findings provide cross‐sectional baseline evidence supporting cautious integration of zoo‐based monitoring into One Health antimicrobial resistance surveillance.

graphic file with name EMI4-18-e70416-g003.webp

1. Introduction

Antimicrobial resistance (AMR) is widely recognised as an environmental as well as a public health problem, shaped by microbial exchange across connected humans, animals and built ecosystems (Larsson et al. 2023). Zoos may contribute to One Health‐oriented AMR surveillance because they bring diverse animal populations into managed settings near staff, visitors and urban environments (Caramujo et al. 2026). However, faecal surveys alone can only indicate carriage in sampled animals and cannot by themselves resolve transmission across the human–animal–environment interface (Nishigaki et al. 2025).

Among enteric bacteria of One Health concern, nontyphoidal Salmonella enterica (NTS) remains a leading cause of gastroenteritis globally and is also an important pathogen for animals (Galán‐Relaño et al. 2023). In managed collections, NTS has been associated with outbreaks in captive populations and has been documented as an infection concern in children's petting zoo settings, reflecting the potential for close contact environments to facilitate exposure (Friedman et al. 1998; Göttling et al. 2022). From an AMR perspective, NTS is also a priority because resistant strains can compromise treatment of invasive infections in vulnerable human populations and complicate veterinary care in valuable captive animals (Marin et al. 2022). Whole‐genome sequencing (WGS) has become central to modern Salmonella surveillance because it enables high‐resolution typing and detection of resistance determinants (Galán‐Relaño et al. 2023). In zoological collections, this genomic lens can help distinguish sporadic introductions from persistent circulation within enclosures or shared management systems, thereby supporting targeted interventions that protect animal health and reduce occupational exposure risks for keepers and veterinary staff (Stenger et al. 2025).

In parallel, Enterococcus spp. provide a complementary indicator for AMR monitoring across One Health systems (Zaidi et al. 2024). Enterococci are widespread in faeces and the environment, tolerate diverse stressors and readily acquire and disseminate resistance traits, making them valuable sentinels of antimicrobial selection pressure in humans, animals and environmental compartments (Zaidi et al. 2024). While zoonotic transmission pathways for resistant enterococci are complex and context‐dependent, their detection in managed animal populations is operationally important for stewardship (choice of empiric therapy) and environmental hygiene within zoo settings (Kim et al. 2024).

Zoological institutions in the UAE play key roles in ex situ conservation, veterinary care and reintroduction‐oriented programmes for conservation‐priority native ungulate species such as the Arabian oryx (Chanyandura et al. 2025). Because conservation‐managed ungulates may be moved between facilities for breeding, veterinary interventions or collection management, systematic screening for zoonotic enteric bacteria and AMR can yield benefits that support animal care, biosecurity and evidence‐based husbandry (Habib et al. 2026). In urban zoos, these benefits are amplified by the proximity of staff and visitors and by the broader environmental connectivity inherent to city settings (Kleespies et al. 2022). Despite the strategic value of such screening for enteric bacteria and their AMR, there remains limited baseline information on NTS and Enterococcus spp. circulating in zoo‐housed, native conservation‐priority ungulates in the UAE and the wider region.

Therefore, in this single‐institution cross‐sectional study we investigated faecal carriage, species/serovar composition and phenotypic AST profiles of NTS and clinically relevant Enterococcus in conservation‐priority captive ungulates. WGS was performed for all recovered NTS isolates and for one phenotypically prioritised Enterococcus faecium isolate. This study highlights the value of integrated phenotypic and genomic approaches for detecting clinically relevant resistance and provides a pilot baseline to inform zoo biosecurity and One Health surveillance in conservation‐managed ungulates.

2. Materials and Methods

2.1. Study Site, Animals and Faecal Sampling

Sampling was undertaken at Al Ain Zoo (https://www.alainzoo.com/; Al Ain City, eastern UAE), a large urban zoological institution that maintains extensive native and exotic collections and supports regional conservation and captive‐breeding programmes (Chanyandura et al. 2025). Study animals were clinically healthy captive ungulates selected through a husbandry‐integrated convenience sampling aligned with daily management schedules and welfare considerations so that procedures could be completed without restraint or disturbance to the animals.

In this study we relied exclusively on freshly voided faeces collected noninvasively from enclosure floors during routine husbandry, with no animal handling beyond standard care. Accordingly, formal animal ethics approval was not required for this component of the study. This design was intended to maximise welfare‐compatible sampling coverage, but it was not a random population survey; therefore, prevalence estimates and host‐species comparisons should be interpreted descriptively. Collection was performed by trained zookeepers during morning rounds. To ensure correct attribution to the defecating individual, samples were taken immediately after observed defecation with the animal visually confirmed at the time of collection. Disposable gloves and sterile single‐use tools were used for each sample, and faeces were transferred into individually labelled sterile containers to reduce the risk of environmental or between‐sample contamination (Habib et al. 2026).

In total, 101 individual ungulates were sampled from an overall source population of approximately 250 animals across the targeted species held at the zoo during the study period. The 101 sampled animals represented approximately 40% of the target source population, but selection depended on animal availability, observed defecation and keeper access during the sampling window. The sampled animals comprised Arabian oryx ( Oryx leucoryx ) (n = 18; 17.9%), scimitar‐horned oryx ( Oryx dammah ) (n = 18; 17.9%), addax ( Addax nasomaculatus ) (n = 21; 20.8%), common eland ( Taurotragus oryx ) (n = 16; 15.8%), Mhorr (dama) gazelle ( Nanger dama mhorr ) (n = 16; 15.8%), dama (Dana) gazelle ( Nanger dama ) (n = 9; 8.9%) and beisa oryx ( Oryx beisa ) (n = 3; 2.9%). This sample set represented the largest feasible number of individuals that could be included during the sampling window while maintaining broad coverage of conservation‐priority ungulate taxa in the collection. The sampling window was conducted over 3 months (September–November 2025). After collection, specimens were held at 4°C and transported on the day of sampling to the Veterinary Public Health Research Laboratory, United Arab Emirates University, for bacteriological testing. All samples were processed within 24 h of collection.

2.2. Isolation and Confirmation of NTS and Enterococcus spp.

Faecal samples (~5 g) from each animal were aseptically transferred into 45 mL of buffered peptone water (BPW; Oxoid, UK) (1:9 w/v) and homogenised. The cultures were incubated at 37°C for 18–24 h for nonselective pre‐enrichment (Sodagari et al. 2020). For Salmonella spp. isolation, the BPW pre‐enrichment cultures were further processed according to standard specifies detection of Salmonella spp. in environmental samples, animal feed and human consumption (ISO (International Organization for Standardization) 2002). After 36 h of pre‐enrichment at 37°C, aliquots (1 mL) were inoculated into Muller–Kauffmann tetrathionate novobiocin (TTN) broth (Neogen, USA) and incubated at 37°C for 48 h. In parallel, 0.1 mL of the pre‐enrichment was inoculated into modified semisolid Rappaport–Vassiliadis (MSRV) medium (Neogen, USA) and incubated at 41.5°C for 48 h. MSRV plates were examined at 24 h for characteristic Salmonella migration (a turbid white zone with radius > 10 mm); plates without visible growth were incubated an additional 24 h. After selective enrichment, aliquots were streaked onto xylose lysine deoxycholate (XLD) agar (Oxoid, UK) and incubated at 37°C for 24 h (ISO (International Organization for Standardization) 2002). Colonies displaying typical Salmonella morphology (e.g., red colonies with black centres on XLD) were picked (one isolate per sample was identified) and purified by subculturing on nutrient agar (37°C, 24 h) (Sodagari et al. 2020).

For Enterococcus spp. isolation, we used direct plating procedures (Habib et al. 2022). Briefly, the homogenate (1:9 w/v) of faecal samples (10−1) underwent further serial dilutions (10−2–10−4) prepared in sterile 0.1% peptone water. From each dilution, 100 μL was plated onto Slanetz–Bartley agar (Neogen, USA) and incubated at 41.5°C for 36–48 h. Colonies characteristic of enterococci (small red or maroon colonies) were selected (one isolate per sample was retained) and subcultured on nutrient agar at 37°C for 24 h to obtain pure cultures (Habib et al. 2022). Because only one representative colony was retained per positive sample, co‐colonising strains or minority resistant subpopulations may not have been detected.

All purified isolates (both presumptive Salmonella and Enterococcus) were then identified by MALDI‐TOF MS using the Autobio MS1000 system (Autobio Diagnostics, Zhengzhou, China). Briefly, single colonies were smeared onto the stainless‐steel target plate, overlaid with α‐cyano‐4‐hydroxycinnamic acid matrix and air‐dried. Spectra were acquired using the manufacturer's software, and identification scores ≥ 9.000 were considered reliable for species‐level assignment. Confirmed isolates were stored at −80°C for subsequent analyses (Park et al. 2021).

2.3. Phenotypic Antimicrobial Susceptibility Testing (AST)

AST for all confirmed Salmonella was performed using the VITEK 2 Compact automated system (bioMérieux, Marcy l'Étoile, France) in accordance with the manufacturer's instructions. The AST‐GN96 card was used, and minimum inhibitory concentration (MIC) values were generated automatically by the instrument software. This panel includes third‐ and fourth‐generation cephalosporins tested both alone and in combination with clavulanic acid, enabling phenotypic detection of extended‐spectrum β‐lactamase (ESBL) production based on clavulanate synergy. AST of Salmonella isolates was extended to 17 antimicrobial agents spanning veterinary‐relevant and clinically important antimicrobial classes used for AMR surveillance. The panel included beta‐lactams, including cephalosporins and imipenem; aminoglycosides; fluoroquinolones/quinolones; tetracycline; phenicols and folate pathway inhibitors. Some agents, such as imipenem, were included as sentinel compounds for surveillance and interpretive context rather than as routine therapeutic options in zoo ungulates.

Following species identification by MALDI‐TOF MS, only E. faecium and Enterococcus faecalis were selected for AST, as these species are the most clinically relevant enterococci in human and veterinary medicine and are commonly associated with multidrug resistance (MDR) and zoonotic transmission potential (Kim et al. 2024; Zaidi et al. 2024). For the selected clinically relevant Enterococcus isolates ( E. faecium and E. faecalis ), susceptibility testing was conducted using the VITEK 2 AST‐P592 card. This panel assessed activity against 10 antimicrobials: ampicillin, erythromycin, ciprofloxacin, high‐level gentamicin, high‐level streptomycin, tigecycline, teicoplanin, tetracycline, vancomycin and linezolid.

For Salmonella, E. faecium and E. faecalis MIC results were interpreted according to the Clinical and Laboratory Standards Institute (CLSI) performance standards (CLSI (Clinical and Laboratory Standards Institute) 2023). Isolates were categorised as MDR when nonsusceptibility was observed to at least one antimicrobial agent in three or more antimicrobial classes, following internationally accepted criteria.

2.4. Whole‐Genome Sequencing Workflow and in Silico Characterisation

Genomic DNA library preparation and Illumina sequencing (2 × 150 bp) were outsourced to Novogene (https://www.novogene.com/; Cambridge, UK) using an Illumina HiSeq platform (Illumina, San Diego, CA, USA). Sequence data were processed in batch using the Solu Genomics platform (Solu Healthcare Inc., Helsinki, Finland; v1.0.600) to support standardised assembly and downstream characterisation (Saratto et al. 2025). The analysis pipeline covered de novo assembly followed by species verification, molecular typing, detection of acquired AMR determinants and resistance‐associated mutations, plasmid reconstruction/typing and virulence gene screening (Saratto et al. 2025). Species assignments were independently validated using Kraken2 v2.1.3 against the PlusPF‐16 reference database (15 October 2025 release; https://github.com/DerrickWood/kraken2). Multilocus sequence types were determined with mlst v2.23.0 using allele and ST definitions from PubMLST (https://github.com/tseemann/mlst). Acquired AMR genes and relevant point mutations were identified using AMRFinderPlus v4.0.23–2025‐07‐16.1 under default settings (https://github.com/ncbi/amr). Plasmid content was characterised using MOB‐suite v3.1.9 (https://github.com/phac‐nml/mob‐suite). Virulence factors are screened using ABRicate v1.0.1 (https://github.com/tseemann/abricate) against the Virulence Factor Database (VFDB; https://www.mgc.ac.cn/VFs/). For Salmonella isolates, in silico serovar prediction was performed using SISTR v1.1.3 (Yoshida et al. 2016). Finally, the genomic sequences were evaluated for predicted pathogenic capacity in humans using the Global Pathogen Analysis Platform via PathogenFinder2, as a supplementary genome‐content screen (Ferrer Florensa et al. 2025). Raw sequencing reads were deposited in the National Center for Biotechnology Information (NCBI) under BioProject PRJNA1438090.

3. Results

3.1. Culture‐Based Detection of NTS and Enterococcus spp.

Results in Table 1 describe the species identity of the retained representative isolate per positive sample rather than the full within‐sample community. Across faecal samples from 101 captive ungulates in Al Ain Zoo, NTS was detected in 4/101 samples (3.9%, 95% CI 1.5%–9.7%), while Enterococcus spp. were detected in 77/101 samples (76.2%, 95% CI 67.1%–83.5%) (Table 1). The four Salmonella isolates comprised serovars Schwarzengrund (n = 2), Kentucky (n = 1) and Chester (n = 1) (Table 1). While the 77 Enterococcus isolates were spread over five species, of which E. faecium (n = 32) and Enterococcus casseliflavus were well represented, each accounted for 41.5% (32/77) of the overall Enterococcus examined (Table 1). Consistent with the study workflow, one isolate per sample was confirmed by MALDI‐TOF MS and species counts therefore equalled the number of positive samples (Table 1). The identified E. faecium (n = 32) and E. faecalis (n = 1) were included in the downstream AST, as these species are the most clinically relevant enterococci in human and veterinary medicine and are commonly associated with zoonotic transmission potential.

TABLE 1.

Culture‐based detection of nontyphoidal Salmonella enterica and Enterococcus spp. in faecal samples from captive ungulates (n = 101) at Al Ain Zoo, United Arab Emirates.

Bacterial group Positive/total (n/N) a Detection%, 95% CI b Serovars/species among isolates (n)
Nontyphoidal Salmonella 4/101 3.9% (1.5%–9.7%) Serovars (n): Schwarzengrund (2); Kentucky (1); Chester (1)
Enterococcus spp. 77/101 76.2% (67.1%–83.5%) Species (n): Enterococcus faecium (32); Enterococcus casseliflavus (32); Enterococcus hirae (9); Enterococcus mundtii (2); Enterococcus avium (1); Enterococcus faecalis (1)
Open in a new tab
a

One isolate per sample was identified by MALDI‐TOF MS; counts equal the number of positive samples.

b

95% confidence interval calculated using the Wilson method.

3.2. Phenotypic Antimicrobial Susceptibility Profiles and MDR Classification

Heatmap summaries are shown in Figure 1, where cell values are presented as S/I/R percentages (susceptible/intermediate/resistant; e.g., 100/0/0) for each animal–organism combination and antimicrobial. Phenotypic AST for Enterococcus focused on 33 isolates ( E. faecium , n = 32; E. faecalis , n = 1) recovered from six host groups (Mhorr (dama) gazelle, Arabian oryx, scimitar‐horned oryx, common eland, dama (Dana) gazelle and addax). Resistance was uncommon and concentrated in a limited subset of agents. Across the panel, the highest resistance frequency was observed for erythromycin (9.1%; 3/33) and ciprofloxacin (9.1%; 3/33). Resistance to ampicillin (3.0%; 1/33), tetracycline (3.0%; 1/33), vancomycin (3.0%; 1/33) and teicoplanin (3.0%; 1/33) was detected at low frequency, while 0% resistance was observed for linezolid, tigecycline and both high‐level aminoglycoside synergy screens (Figure 1).

FIGURE 1.

FIGURE 1

Open in a new tab

Heatmap of phenotypic antimicrobial susceptibility profiles of (A) Enterococcus faecium and Enterococcus faecalis isolates and (B) nontyphoidal Salmonella enterica isolates recovered from captive ungulates at Al Ain Zoo, United Arab Emirates. Cell values are shown as raw S/I/R counts, where S, susceptible; I, intermediate; R, resistant; for example, ‘S 2/2’ indicates that 2 of 2 isolates were susceptible. Row labels indicate the number of isolates tested for each host–bacterial group combination. AMC, amoxicillin–clavulanic acid; AMP, ampicillin; CEF, cefalotin; CFP, cefoperazone; CIP, ciprofloxacin; CQN, cefquinome; EFT, ceftiofur; ENR, enrofloxacin; ERY, erythromycin; FFC, florfenicol; FLU, flumequine; GEN, gentamicin; GEN‐HL, high‐level gentamicin; IPM, imipenem; LEX, cefalexin; LZD, linezolid; MAR, marbofloxacin; NEO, neomycin; STR‐HL, high‐level streptomycin; SXT, trimethoprim–sulfamethoxazole; TCC, ticarcillin–clavulanic acid; TEC, teicoplanin; TET, tetracycline; TGC, tigecycline; VAN, vancomycin.

Phenotypic AST was done for the four recovered NTS isolates (serovars Schwarzengrund, n = 2; Kentucky, n = 1; Chester, n = 1). A shared nonsusceptibility pattern was observed across all four isolates driven by uniform resistance to cefalexin (100%; 4/4), cefalotin (100%; 4/4) and gentamicin (100%; 4/4), together with florfenicol nonsusceptibility (100%; 4/4) (intermediate in 3/4 and resistant in 1/4) (Figure 1). Additional resistance was confined to a single isolate (serovar Kentucky), which displayed resistance to multiple antimicrobial classes including ampicillin, neomycin, flumequine, enrofloxacin, marbofloxacin, tetracycline, florfenicol and trimethoprim/sulfamethoxazole and intermediate results for amoxicillin/clavulanate and ticarcillin/clavulanate (Figure 1).

Applying the study MDR definition (resistance to > 1 agent in ≥ 3 antimicrobial groups), none (0/33) of the examined Enterococcus isolates met MDR criteria based on phenotypic AST. Nevertheless, one E. faecium isolate (from scimitar‐horned oryx) showed concurrent glycopeptide resistance (vancomycin and teicoplanin) alongside ampicillin resistance, and this isolate was retained for WGS as the phenotypically most clinically relevant profile within the Enterococcus dataset. On the other hand, 4/4 (100%) NTS isolates met MDR criteria, reflecting the combined nonsusceptibility across cephalosporins, aminoglycosides and phenicols in all isolates, with substantially broader class‐level nonsusceptibility in the serovar Kentucky isolate (Figure 1).

3.3. Whole‐Genome Sequencing Characterisation

WGS was performed for all NTS isolates and for a phenotypically prioritised E. faecium isolate (Table 2). The sequenced E. faecium isolate (ZEF) was assigned to ST22 and carried AMR determinants including aac(6′)‐I (aminoglycoside resistance), msr(C) (macrolide resistance) and a liaR_E75K substitution (associated with reduced susceptibility to daptomycin) (Table 2). Two plasmids were reconstructed, including a 50,719 bp plasmid (MOB cluster AD907, relaxase MOBP, conjugative) and a 23,029 bp plasmid (MOB cluster AB759, relaxase MOBP, mobilisable) (Table 2).

TABLE 2.

Whole‐genome sequencing based characterisation of all nontyphoidal Salmonella enterica isolates and a multidrug‐resistant Enterococcus faecium from captive ungulates at Al Ain Zoo, United Arab Emirates.

Isolate‐serovar/species Short‐read quality Sequence type Antimicrobial resistance determinants Plasmids
Size (bp) MOB cluster Relaxase types Putative mobility
ZEF‐ E. faecium Mean read length: 149.9 bp ST22 aac(6′)‐I, liaR_E75K, msr(C) 50,719 AD907 MOBP Conjugative
23,029 AB759 MOBP Mobilisable
Coverage: 665×
N50: 231 kbp
ZS1‐Salmonella serovar Schwarzengrund Mean read length: 150.0 bp ST241 — a 54,608 AC578 MOBF Conjugative
Coverage: 356×
N50: 337 kbp
ZS2‐Salmonella serovar Kentucky Mean read length: 150.0 bp ST198 aadA1, aadA2, aph(3′)‐Ia, bla TEM‐57, dfrA12, sul1, sul3, tet(A), cmlA1, qacEdelta1, qacL, parC_S80I, gyrA_D87G, gyrA_S83F 21,304 AC247 — Nonmobilisable
Coverage: 520×
10,841 AG658 — Nonmobilisable
N50: 645 kbp
5865 AA525 MOBP Mobilisable
4345 AA575 MOBQ Mobilisable
4221 AA579 MOBQ Mobilisable
ZS3‐Salmonella serovar Schwarzengrund Mean read length: 150.0 bp ST241 — 54,608 AC578 MOBF Conjugative
Coverage: 307×
N50: 425 kbp
ZS4‐Salmonella serovar Chester Mean read length: 150.0 bp ST2063 — — — — —
Coverage: 404×
N50: 587 kbp
Open in a new tab
a

None identified.

For NTS, both Schwarzengrund isolates (ZS1 and ZS3) belonged to ST241, and each carried a 54,608 bp conjugative plasmid (MOB cluster AC578, relaxase MOBF) (Table 2). The Kentucky isolate (ZS2) was ST198 and harboured a broad resistome including aminoglycoside, β‐lactam, sulfonamide/trimethoprim, tetracycline, phenicol and quinolone resistance‐associated determinants and mutations, and quaternary ammonium (disinfectant)‐associated genes (Table 2). Multiple plasmids were reconstructed for ZS2, including two nonmobilisable plasmids and three mobilisable plasmids (Table 2). The Chester isolate (ZS4) was assigned to ST2063 and did not show plasmid reconstruction entries (Table 2).

3.4. Virulence Gene Content and PathogenFinder‐Predicted Human Pathogenic Capacity

Across the four NTS genomes, the number of detected virulence‐associated genes ranged from 143 to 148 (Figure 2). PathogenFinder scores were consistently high (0.9597 for ZS1 Schwarzengrund; 0.9601 for ZS2 Kentucky; 0.9639 for ZS3 Schwarzengrund; 0.963 for ZS4 Chester), all exceeding the > 0.5 threshold used to indicate predicted human pathogenic capacity (Figure 2). Comparative profiling identified a set of variable virulence genes that differed among isolates (Figure 2). Across the four NTS isolates, 26 virulence genes were variable (present in a subset rather than all genomes), with the Kentucky and Chester genomes showing a larger share of the variable repertoire than the two Schwarzengrund genomes (Figure 2).

FIGURE 2.

FIGURE 2

Open in a new tab

Virulence gene content and predicted human pathogenic potential of four nontyphoidal Salmonella isolates recovered from captive ungulates at Al Ain Zoo, United Arab Emirates. The upper panel shows the total number of virulence‐associated genes detected per isolate, with the PathogenFinder score (predicted human pathogenic capacity) displayed above each bar. The lower panel presents a binary heatmap of the distribution of variable virulence genes across isolates, indicating presence (blue) or absence (white) across isolates to highlight differences in gene repertoires.

For the sequenced E. faecium isolate, virulence screening identified only two virulence‐associated genes (acm and scm; collagen adhesin/binding precursors) (Table S1). The PathogenFinder score for this E. faecium genome was 0.6138 (0.0441).

Functional annotation of virulence genes shared (n = 134) among the NTS genomes indicated representation across adhesion/colonisation and secretion‐associated categories, including fimbrial colonisation (nine genes), curli/biofilm‐associated determinants (eight genes), SPI‐1 structural components (21 genes) and effectors/regulators (14 genes), SPI‐2 secretion apparatus genes (22 genes) and intracellular survival effectors (10 genes), global regulators (seven genes), iron acquisition (two genes) and additional conserved virulence‐associated genes (41 genes) (Table S1). For Enterococcus, the virulence complements in the sequenced isolate comprised acm and scm (two genes), categorised as collagen adhesin/binding precursors (Table S1).

4. Discussion

AMR is increasingly recognised as a One Health problem shaped by interactions across human, animal and environmental systems (Larsson et al. 2023). In the Middle East, conservation‐managed ungulates are central to biodiversity, and zoos provide a managed setting where animal health, husbandry and environmental hygiene intersect (Low Cunningham 2004). In the studied urban zoological environment in the UAE, we detected NTS in a small proportion of clinically healthy ungulates alongside frequent carriage of Enterococcus spp. All NTS isolates met the study MDR definition, and the Salmonella serovar Kentucky isolate showed the broadest nonsusceptibility and resistome. While phenotypic resistance among Enterococcus isolates was generally uncommon, we identified a glycopeptide‐resistant E. faecium isolate with a daptomycin‐associated genomic determinant. WGS resolved serovar/species composition, plasmid architectures and predicted high human‐pathogenic potential for the NTS genomes, providing a regional baseline for conservation‐priority zoo ungulates.

Within the examined ungulates in the present study setting, low NTS detection alongside high Enterococcus recovery was consistent with the contrasting ecology of these taxa: NTS carriage can be intermittent and exposure‐driven, whereas enterococci are common commensals that persist under diverse conditions (Rukambile et al. 2019). Subclinical Salmonella carriage has been reported in other zoo collections, including a Mexican survey documenting diverse nontyphoidal serovars (Silva‐Hidalgo et al. 2012) and a South African national zoo study detecting Salmonella in captive animals (Mtshali 2012). Similar faecal surveys in free‐roaming wildlife in South Korea have also detected Salmonella (Mafizur et al. 2024). Nevertheless, the most consequential signal in our dataset was MDR in all NTS isolates recovered from asymptomatic hosts. In conservation‐managed herds, such resistance can narrow therapeutic options if clinical salmonellosis occurs and can increase the operational consequences of outbreaks (Marin et al. 2022). The identification of Salmonella serovar Kentucky ST198 was noteworthy because this lineage has been linked to international dissemination of MDR strains and high‐level fluoroquinolone resistance in human surveillance (Mohamed et al. 2025). Although we did not infer sources or transmission, the broader resistome in our Salmonella ST198 isolate aligned with this high‐concern lineage (Le Hello et al. 2013).

The Enterococcus findings complemented the NTS results by reflecting commensal carriage in the same managed environment. Enterococci are used as indicators of faecal contamination and antimicrobial selection because they are abundant, resilient and genetically adaptable (Zaidi et al. 2024). Resistance profiles in wildlife and captivity have varied, including reports from a zoo collection (Kim et al. 2024), captive parrots (de Freitas et al. 2018) and free‐living primates (Grassotti et al. 2018). Genomic work in free‐living birds supports environmental reservoirs for potentially pathogenic Enterococcus lineages (Kwit et al. 2023). In our data, no E. faecium / E. faecalis isolates met the phenotypic MDR definition, but a single glycopeptide‐resistant E. faecium isolate was clinically notable, based on its phenotypic resistance to priority last‐line antimicrobials (glycopeptides) (van Groesen et al. 2022). WGS assigned it to ST22, which has been identified before in E. faecium from healthy humans and free‐living birds (Kwit et al. 2023; Tedim et al. 2024). Additionally, WGS identified an amino‐acid substitution in liaR_E75K associated with reduced susceptibility to daptomycin, a last‐line agent used against severe MDR Gram‐positive infections (Zeng et al. 2022). Because daptomycin was not included in the phenotypic AST panel, the liaR_E75K substitution should be interpreted only as a genomic alert requiring confirmatory daptomycin MIC testing. Similarly, the glycopeptide phenotype should be confirmed by reference MIC testing and targeted evaluation of known glycopeptide‐resistance determinants before mechanistic interpretation. Such findings underscore the importance of integrating genomic analyses with phenotypic testing to better interpret AMR mechanisms and their potential clinical and ecological implications.

WGS is increasingly accepted as a gold‐standard approach for Salmonella serovar determination, offering higher resolution and overcoming several limitations associated with conventional serotyping methods (Yoshida et al. 2016). In the present study in a zoo setting, WGS added resolution beyond AST by placing each NTS isolate within recognised serovar and sequence types and characterising mobile genetic elements. The two isolates belonging to Salmonella serovar Schwarzengrund carried a conjugative plasmid and were assigned ST241 that has been described in poultry‐associated Salmonella Schwarzengrund with evidence of dissemination in broiler systems (Felix et al. 2025). The serovar Chester isolate was ST2063, a sequence type reported in Salmonella from geographically structured clusters and was associated with travel‐associated salmonellosis outbreaks (Fonteneau et al. 2017; Siira et al. 2019). These contextual reports do not imply direct linkage to our zoo setting but illustrate how WGS enables standardised comparisons across sectors.

Virulence profiling and PathogenFinder predictions further supported One Health relevance of the present study findings. All NTS genomes carried large virulence gene repertoires with conserved SPI‐1/SPI‐2 machinery, and PathogenFinder scores were uniformly high (> 0.95), consistent with predicted human‐pathogenic potential based on genome content (Ferrer Florensa et al. 2025). Such predictions should be interpreted as compatibility with human pathogenicity rather than evidence of transmission. For the sequenced E. faecium , only the collagen adhesin/binding genes acm and scm were detected, and the PathogenFinder score was modestly above the 0.5 threshold (0.6138 [0.0441]) (Ferrer Florensa et al. 2025). Overall, these findings reinforced the urban zoo as a sentinel environment at the human–animal–environment interface. Culture‐based detection, standardised AST and WGS offered complementary information for zoo biosecurity by identifying subclinical carriages in conservation animals and defining MDR and mobile‐element features relevant to management.

Several limitations should be noted. The study was cross‐sectional and limited to a single institution, restricting inference about temporal dynamics or persistence. The small number of NTS isolates constrained epidemiological conclusions and host‐level comparisons. The Enterococcus genomic component was limited to one phenotypically prioritised E. faecium isolate and therefore cannot support conclusions about the genomic diversity or resistome of Enterococcus spp. in the sampled population. Environmental, feed, water and human sampling were not included, preventing possible source attribution. Finally, the one‐isolate‐per‐sample approach improved comparability but almost certainly reduced detection of within‐sample diversity, mixed serovars/species and co‐occurring resistance profiles. Despite these limitations, this study represents the first investigation in the UAE (and to our knowledge the wider Middle East) focusing on AMR and genomic characterisation of these specific groups of enteric bacteria in native conservation‐priority ungulates. The study was conducted within one of the region's leading zoological institutions for wildlife conservation; as such, the findings provide an important baseline for future surveillance and One Health research in zoological and conservation settings.

5. Conclusion

In conclusion, conservation‐priority ungulates housed in an urban UAE zoo carried both NTS and Enterococcus spp., including MDR NTS lineages and a phenotypically notable E. faecium isolate with glycopeptide nonsusceptibility that warrants confirmatory testing and further genomic investigation. These findings support the usefulness of targeted zoo‐based faecal surveillance as one component of One Health AMR monitoring, while broader longitudinal sampling of animals, environment, feed, water, staff‐contact points and visitors would be required to evaluate transmission pathways.

Author Contributions

Ihab Habib: conceptualization, methodology, formal analysis, supervision, writing – original draft, writing – review and editing, funding acquisition. Moneeb A. Qablan: data curation, supervision, validation, resources, writing – review and editing. Fatema Aaref Alzaabi: writing – review and editing, investigation, validation, formal analysis. Fatema Rashed Alnuaimi: investigation, validation, formal analysis, writing – review and editing. Ghalya Ahmed Almemari: investigation, validation, formal analysis, writing – review and editing. Glindya Bhagya Lakshmi: supervision, project administration, validation, methodology, writing – review and editing. Safeya Al Naqbi: data curation, resources, validation, writing – review and editing. Fatima Al Shamsi: data curation, resources, validation, writing – review and editing. Dhanya Vijay: data curation, resources, validation, writing – review and editing. John Mhongovoyo: resources, project administration, validation, writing – review and editing. Sultan Khalifa Alkaabi: resources, project administration, validation, writing – review and editing. Amna Mana Alotaiba: resources, project administration, validation, writing – review and editing, funding acquisition.

Funding

This work was supported by the United Arab Emirates University (G00005128, G00005130).

Ethics Statement

No animal ethics approval was required, as faecal samples were collected noninvasively during routine husbandry without animal handling or experimental intervention.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: Functional categorisation of the virulence genes shared among the genomes of nontyphoidal Salmonella isolates, and multidrug‐resistant Enterococcus faecium from captive ungulates at Al Ain Zoo, United Arab Emirates.

EMI4-18-e70416-s001.docx (17.5KB, docx)

Acknowledgements

The authors acknowledge Al Ain Zoo and its animal keepers, veterinary services and animal management team for facilitating sample collection and for their collaboration and commitment to best practices in animal care and conservation. Funding for research consumables and all laboratory analyses was provided by the United Arab Emirates University (UAEU) Sustainable Development Goals (SDGs) Research Program under Grant Code G00005128, awarded to Dr. Ihab Habib. Field sample collection was supported by the UAEU‐SDGs Research Program under Grant Code G00005130, awarded to Dr. Moneeb A. Qablan.

During the preparation of this manuscript, the authors used OpenAI ChatGPT, GPT‐5.5 Thinking (large language model) and Grammarly, 2026 version/platform (AI writing support platform), to assist only with language‐related editing. These tools were used in 2026 to support sentence structure improvement, grammar correction, stylistic refinement, rephrasing and clarity enhancement. AI assistance was applied only to the textual content of the manuscript, including the Abstract, Introduction, Methods, Results, Discussion and Conclusion sections, where needed for language polishing. The AI tools were not used to generate original research ideas, develop the study design, collect or analyse data, interpret results, create or modify figures or tables, generate references or alter the scientific content or conclusions of the manuscript. The authors directed the use of these tools, reviewed and edited all AI‐assisted suggestions, verified the accuracy and integrity of the final manuscript, and take full responsibility for the content of the publication.

Data Availability Statement

The sequencing data and assembled genomes generated in this study have been publicly deposited in the NCBI database under BioProject accession number PRJNA1438090.

References

  1. Caramujo, S. , Abreu R., Pereira G., et al. 2026. “Antimicrobial Resistance in Escherichia coli From Captive Wild Felids: Associations With Host and Management Factors.” Veterinary Sciences 13, no. 2: 124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Chanyandura, A. , Alotaiba A. M., Alqahtani H. A., and Alhajeri M. M.. 2025. “Ensuring Species' Survival and Longevity: The Role of Modern Zoos.” Animal Reports 1, no. 2: 145–168. [Google Scholar]
  3. CLSI (Clinical and Laboratory Standards Institute) . 2023. Performance Standards for Antimicrobial Susceptibility Testing. 33rd ed. Clinical and Laboratory Standards Institute. [Google Scholar]
  4. de Freitas, A. D. A. R. , Faria A. R., Pinto T. D. C. A., et al. 2018. “Distribution of Species and Antimicrobial Resistance Among Enterococci Isolated From the Fecal Microbiota of Captive Blue‐Fronted Parrot ( Amazona aestiva ) in Rio de Janeiro, Brazil.” Science of the Total Environment 615: 1428–1437. [DOI] [PubMed] [Google Scholar]
  5. Felix, M. A. , Han J., Khajanchi B. K., Sanad Y. M., Zhao S., and Foley S. L.. 2025. “ Salmonella enterica Serovar Schwarzengrund: Distribution, Virulence, and Antimicrobial Resistance.” Microorganisms 13, no. 1: 92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Ferrer Florensa, A. , Almagro Armenteros J. J., Kaas R. S., et al. 2025. “Whole‐Genome Prediction of Bacterial Pathogenic Capacity on Novel Bacteria Using Protein Language Models, With PathogenFinder2.” Bioinformatics 42, no. 5: btag129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Fonteneau, L. , Da Silva N. J., Fabre L., et al. 2017. “Multinational Outbreak of Travel‐Related Salmonella Chester Infections in Europe, Summers 2014 and 2015.” Eurosurveillance 22, no. 7: 30463. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Friedman, C. R. , Torigian C., Shillam P. J., et al. 1998. “An Outbreak of Salmonellosis Among Children Attending a Reptile Exhibit at a Zoo.” Journal of Pediatrics 132, no. 5: 802–807. [DOI] [PubMed] [Google Scholar]
  9. Galán‐Relaño, Á. , Valero Díaz A., Huerta Lorenzo B., et al. 2023. “ Salmonella and Salmonellosis: An Update on Public Health Implications and Control Strategies.” Animals: An Open Access Journal From MDPI 13, no. 23: 3666. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Göttling, J. , Heckel J. O., Hotzel H., et al. 2022. “Zoonotic Bacteria in Clinically Healthy Goats in Petting Zoo Settings of Zoological Gardens in Germany.” Zoonoses and Public Health 69, no. 4: 333–343. [DOI] [PubMed] [Google Scholar]
  11. Grassotti, T. T. , de Angelis Zvoboda D., da Fontoura Xavier Costa L., et al. 2018. “Antimicrobial Resistance Profiles in Enterococcus spp. Isolates From Fecal Samples of Wild and Captive Black Capuchin Monkeys ( Sapajus nigritus ) in South Brazil.” Frontiers in Microbiology 9: 2366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Habib, I. , Lakshmi G. B., Mohamed M. Y. I., Ghazawi A., Khan M., and Li D.. 2022. “Enumeration, Antimicrobial Resistance, and Virulence Genes Screening of Enterococcus spp. Isolated From Retail Chicken Carcasses in The United Arab Emirates.” Foodborne Pathogens and Disease 19, no. 9: 590–597. [DOI] [PubMed] [Google Scholar]
  13. Habib, I. , Qablan M. A. M., Alzaabi F. A., et al. 2026. “Captive Ungulates as Sentinels of Antimicrobial Resistance: Genomic and Phenotypic Characterization of Extended‐Spectrum Beta‐Lactamase‐Producing Escherichia coli at a United Arab Emirates Urban Zoo.” Frontiers in Veterinary Science 13: 1815262. 10.3389/fvets.2026.1815262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. ISO (International Organization for Standardization) . 2002. ISO 6579:2002. Microbiology of Food and Animal Feeding Stuffs—Horizontal Method for the Detection of Salmonella spp. ISO. [Google Scholar]
  15. Kim, M. , Kim M., Yeo Y.‐G., Lee Y.‐T., and Han J.‐I.. 2024. “Antimicrobial Resistance of Commensal Escherichia coli and Enterococcus faecalis Isolated From Clinically Healthy Captive Wild Animals in Seoul Zoo.” Frontiers in Veterinary Science 10: 1283487. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kleespies, M. W. , Feucht V., Becker M., and Dierkes P. W.. 2022. “Environmental Education in Zoos—Exploring the Impact of Guided Zoo Tours on Connection to Nature and Attitudes Towards Species Conservation.” Journal of Zoological and Botanical Gardens 3, no. 1: 56–68. [Google Scholar]
  17. Kwit, R. , Zając M., Śmiałowska‐Węglińska A., et al. 2023. “Prevalence of Enterococcus spp. and the Whole‐Genome Characteristics of Enterococcus faecium and Enterococcus faecalis Strains Isolated From Free‐Living Birds in Poland.” Pathogens 12, no. 6: 836. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Larsson, D. G. J. , Gaze W. H., Laxminarayan R., and Topp E.. 2023. “AMR, One Health and the Environment.” Nature Microbiology 8, no. 5: 754–755. [DOI] [PubMed] [Google Scholar]
  19. Le Hello, S. , Bekhit A., Granier S. A., et al. 2013. “The Global Establishment of a Highly‐Fluoroquinolone Resistant Salmonella enterica Serotype Kentucky ST198 Strain.” Frontiers in Microbiology 4: 395. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Low Cunningham, P. 2004. “Checklist and Status of the Terrestrial Mammals From The United Arab Emirates.” Zoology in the Middle East 33, no. 1: 7–20. [Google Scholar]
  21. Mafizur, R. M. , Sangjin L., and Chul P. Y.. 2024. “Prevalence of Salmonella spp. and Escherichia coli in the Feces of Free‐Roaming Wildlife Throughout South Korea.” PLoS One 19, no. 2: e0281006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Marin, C. , Martín‐Maldonado B., Cerdà‐Cuéllar M., et al. 2022. “Antimicrobial Resistant Salmonella in Chelonians: Assessing Its Potential Risk in Zoological Institutions in Spain.” Veterinary Sciences 9, no. 6: 264. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Mohamed, M. Y. I. , Khalifa H. O., and Habib I.. 2025. “Food Pathways of Salmonella and Its Ability to Cause Gastroenteritis in North Africa.” Food 14, no. 2: 253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Mtshali, K. 2012. “Detection of Salmonella, Clostridium Perfringens and Escherichiacoli From Fecal Samples of Captive Animals at the National Zoological Gardens of SouthAfrica.” African Journal of Microbiology Research 6: 3662–3666. 10.5897/AJMR12.107. [DOI] [Google Scholar]
  25. Nishigaki, A. , Arden K., and Frosini S. M.. 2025. “Antimicrobial Resistance in Petting Zoo Animals in the United Kingdom.” Applied Microbiology 5, no. 4: 115. [Google Scholar]
  26. Park, J. H. , Jang Y., Yoon I., Kim T. S., and Park H.. 2021. “Comparison of Autof ms1000 and Bruker Biotyper MALDI‐TOF MS Platforms for Routine Identification of Clinical Microorganisms.” BioMed Research International 2021, no. 1: 6667623. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Rukambile, E. , Sintchenko V., Muscatello G., Kock R., and Alders R.. 2019. “Infection, Colonization and Shedding of Campylobacter and Salmonella in Animals and Their Contribution to Human Disease: A Review.” Zoonoses and Public Health 66, no. 6: 562–578. [DOI] [PubMed] [Google Scholar]
  28. Saratto, T. , Visuri K., Lehtinen J., Ortega‐Sanz I., Steenwyk J. L., and Sihvonen S.. 2025. “Solu: A Cloud Platform for Real‐Time Genomic Pathogen Surveillance.” BMC Bioinformatics 26, no. 1: 12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Siira, L. , Naseer U., Alfsnes K., Hermansen N. O., Lange H., and Brandal L. T.. 2019. “Whole Genome Sequencing of Salmonella Chester Reveals Geographically Distinct Clusters, Norway, 2000 to 2016.” Eurosurveillance 24, no. 4: 1800186. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Silva‐Hidalgo, G. , Ortiz‐Navarrete V. F., Alpuche‐Aranda C. M., et al. 2012. “Non‐Typhi Salmonella Serovars Found in Mexican Zoo Animals.” Research in Veterinary Science 93, no. 3: 1132–1135. [DOI] [PubMed] [Google Scholar]
  31. Sodagari, H. R. , Habib I., Whiddon S., et al. 2020. “Occurrence and Characterization of Salmonella Isolated From Table Egg Layer Farming Environments in Western Australia and Insights Into Biosecurity and Egg Handling Practices.” Pathogens 9, no. 1: 56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Stenger, B. L. , Evans D. R., Gefroh S. J., Breuer A., Pecoraro H. L., and Maddock K. J.. 2025. “Spots, Stripes, and Strains: A Case Report of Multidrug‐Resistant Salmonella Newport in Exotic Felids.” ASM Case Reports 2: e00098‐25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Tedim, A. P. , Almeida‐Santos A. C., Lanza V. F., et al. 2024. “Bacteriocin Distribution Patterns in Enterococcus faecium and Enterococcus lactis: Bioinformatic Analysis Using a Tailored Genomics Framework.” Applied and Environmental Microbiology 90, no. 10: e01376‐24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. van Groesen, E. , Innocenti P., and Martin N. I.. 2022. “Recent Advances in the Development of Semisynthetic Glycopeptide Antibiotics: 2014–2022.” ACS Infectious Diseases 8, no. 8: 1381–1407. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Yoshida, C. E. , Kruczkiewicz P., Laing C. R., et al. 2016. “The Salmonella in Silico Typing Resource (SISTR): An Open Web‐Accessible Tool for Rapidly Typing and Subtyping Draft Salmonella Genome Assemblies.” PLoS One 11, no. 1: e0147101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Zaidi, S. E. Z. , Zaheer R., Zovoilis A., and Mcallister T. A.. 2024. “Enterococci as a One Health Indicator of Antimicrobial Resistance.” Canadian Journal of Microbiology 70, no. 8: 303–335. [DOI] [PubMed] [Google Scholar]
  37. Zeng, W. , Feng L., Qian C., et al. 2022. “Acquisition of Daptomycin Resistance by Enterococcus faecium Confers Collateral Sensitivity to Glycopeptides.” Frontiers in Microbiology 13: 815600. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1: Functional categorisation of the virulence genes shared among the genomes of nontyphoidal Salmonella isolates, and multidrug‐resistant Enterococcus faecium from captive ungulates at Al Ain Zoo, United Arab Emirates.

EMI4-18-e70416-s001.docx (17.5KB, docx)

Data Availability Statement

The sequencing data and assembled genomes generated in this study have been publicly deposited in the NCBI database under BioProject accession number PRJNA1438090.

Articles from Environmental Microbiology Reports are provided here courtesy of Wiley

RESOURCES