Unravelling the Pathotype, Biofilm, Virulome and Resistome Profile of MultiDrug‐Resistant Bacteria Isolated From Cattle Faeces Using Whole Genome Sequence Analysis
Tesleem Olatunde Abolarinwa, Daniel Jesuwenu Ajose, Bukola Opeyemi Oluwarinde, Akamu Jude Ewunkem, Peter Kotsoana Montso, Jean Engohang-Ndong, Todd Riley Callaway, Omolola Esther Fayemi, Adeyemi Oladapo Aremu, Collins Njie Ateba
TL;DR
This study identifies multidrug-resistant bacteria in cattle feces and analyzes their potential to cause disease and spread antibiotic resistance.
Contribution
The study provides new insights into the virulence and resistance profiles of multidrug-resistant bacteria from asymptomatic cattle.
Findings
Three multidrug-resistant bacteria isolates (DEC_NWU, DVC_NWU, DSS_NWU) were identified from cattle feces.
The isolates were confirmed as E. coli, V. cholerae, and S. enterica with high similarity to human pathogenic strains.
DSS_NWU had the highest number of virulence genes (65), indicating significant disease potential.
Abstract
The high mortality and morbidity resulting from diarrhoeal cases worldwide are associated with the increasing incidence of antimicrobial resistance (AMR) and represent a serious public health concern. Cattle are a major reservoir of AMR organisms, and faecal shedding may facilitate their transmission into the food chain. This study examined the pathotype, biofilm, virulome and resistome profiles of bacteria isolated from cattle faeces using whole genome sequencing (WGS). Asymptomatic cattle faecal samples (n = 269) were analysed, and three isolates identified as multidrug‐resistant and biofilm‐forming bacteria were sequenced. In this study, we successfully isolated bacteria from cattle faecal samples, and the isolates DEC_NWU, DVC_NWU and DSS_NWU were phenotypically confirmed as multidrug‐resistant and strong biofilm formers. WGS analysis confirmed DEC_NWU, DVC_NWU and DSS_NWU to have…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
FIGURE 2
FIGURE 3
FIGURE 4
FIGURE 5
FIGURE 6
FIGURE 7
FIGURE 8| Bacteria | Target gene | Primer sequence | PCR condition | Amplicon size (bp) | Reference |
|---|---|---|---|---|---|
| All bacterial isolates | 16S rRNA | 27F = AGAGTTTGATCATGGCTCAG | Denaturation at 94°C for 5 min, annealing at 57°C for 30 s, elongation at 72°C for 1 min and 30 cycles | 1420 | [ |
| 1492R = GGTACCTTGTTACGACTT | |||||
|
|
| F=CTGGTATCAGCGCGAAGTCT | Denaturation at 94°C for 4 min, annealing at 47°C for 30 s, elongation at 72°C for 1 min and 25 cycles | 600 | [ |
| R = AGCGGGTAGATATCACACCTC | |||||
|
|
| F = GTGAAATTATCGCCACGTTCGGGCAA | Denaturation at 94°C for 4 min, annealing at 55°C for 30 s, elongation at 72°C for 1 min and 30 cycles | 284 | [ |
| R = TCATCGCACCGTCAAAGGAACC | |||||
|
|
| F = AAGACCTCAACTGGCGGTA | Denaturation at 94°C for 3 min, annealing at 50°C for 30 s, elongation at 72°C for 1 min and 30 cycles | 248 | [ |
| R = GAAGTGTTAGTGATCGCCAGAGT | |||||
| Genome content | Isolate | ||
|---|---|---|---|
| DEC_NWU | DVC_NWU | DSS_NWU | |
| Genome length | 4803571 bp | 4499945 bp | 5,374,783 bp |
| Contigs N50 | 183,022 bp | 659,577 bp | 246,095 bp |
| Number of contigs | 99 | 38 | 146 |
| Contigs L50 | 9 | 2 | 7 |
| Number of protein‐coding sequences (CDS) | 4758 | 4215 | 5624 |
| Species |
|
|
|
| Sequence read archive (SRA) | SRR23903972 | SRR25083111 | SRR23935060 |
| Accession number | |||
| Virulence factor class | Isolate the virulence factor | ||
|---|---|---|---|
| DEC_NWU | DVC_NWU | DSS_NWU | |
| Adherence | CFA/I fimbriae ( | Flagella L‐ring protein ( | Afimbrial adhesin ( |
| Iron uptake | Achromobactin biosynthesis and transport ( | Periplasmic binding protein‐dependent ABC transport systems ( | Outer membrane ( |
| Invasion | Invasion of brain endothelial cells ( | Flagella ( | Signal transduction histidine protein kinase ( |
| Secretion system | ACE Type VI secretion system ( | EPS Type II secretion system ( | Type III secretion system of Salmonella pathogenicity island 1 encode ( |
| Toxin | Haemolysin E/cytolysin A ( | Heat‐stable cytotonic enterotoxin ( | Typhoid toxin ( |
| Drug class | DEC_NWU resistance gene | DVC_NWU resistance gene | DSS_NWU resistance gene |
|---|---|---|---|
| Aminocoumarin |
|
|
|
| Aminoglycoside |
|
|
|
| Carbapenem |
|
|
|
| Cephalosporin |
|
|
|
| Fluoroquinolone |
|
|
|
| Glycopeptide |
|
|
|
| Macrolide |
|
|
|
| Penam |
|
|
|
| Peptide antibiotic |
|
|
|
| Phenicol |
|
|
|
| Phosphonic acid |
|
|
|
| Rifamycin |
|
|
|
| Tetracycline |
|
|
|
- —North-West University10.13039/501100005274
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsBacterial biofilms and quorum sensing · Vibrio bacteria research studies · Milk Quality and Mastitis in Dairy Cows
1. Introduction
The human health burden associated with foodborne complications has made food safety an important global issue requiring serious attention [1]. The presence of pathogens in food products has resulted in the recall of products, especially those that are of animal origin, as such operational standards play a significant role in determining the quality of finished products [1, 2]. Food contamination by pathogens can occur at any stage along the farm‐to‐table continuum from environmental, animal or human sources, resulting in foodborne disease such as diarrhoea [3].
Diarrhoea is one of the most frequent foodborne illnesses resulting from the consumption of contaminated food in both developing and developed countries [4]. Globally, diarrhoea is the third and fifth leading cause of disability‐adjusted life years (DALYs) in children under 5 years old and all ages, respectively [5, 6]. The magnitude of the problems caused by diarrhoea is illustrated by the large proportion of outbreaks that account for more than 1.6 million deaths annually globally [7, 8]. The morbidity and mortality resulting from diarrhoeal cases are worsened due to an increasing incidence of antimicrobial resistance (AMR) among diarrhoeal pathogens, which is now categorized as a global health problem. In 2019, drug‐resistant bacterial infections were directly responsible for an estimated 1.27 million human deaths worldwide, with a further 4.95 million deaths associated with AMR, underscoring its substantial public health impact [9]. The One‐Health perspective is crucial, as resistant bacteria and resistance genes circulate among humans, animals and the environment, facilitated by antimicrobial use across sectors [10–12].
Various pathogens, including viruses, bacteria, parasites and fungi, are known to cause diarrhoea in humans and animals [6, 13]. Enteric bacteria, including Campylobacter jejuni, Escherichia coli, Vibrio cholerae, Salmonella and Shigella species, are among the notable causative agents of diarrhoea [6, 13, 14]. Virulence factors related to the pathogenicity of diarrhoeal pathogens are numerous, ranging from those that facilitate colonization and invasion to toxin production [15]. These virulence determinants also include adhesins, toxins, iron acquisition factors, lipopolysaccharides, polysaccharide capsules and secretion systems, which are usually encoded on pathogenicity islands, plasmids and other mobile genetic elements [15].
Several outbreaks of diarrhoea resulting from the consumption of undercooked food products previously contaminated with faecal matter have been reported worldwide [16, 17]. More than 70% of all diarrhoea outbreak cases reported between 2015 and 2019 were linked to contaminated foods of animal origin [17, 18]. Major food animals, especially cattle, represent the main reservoir of enteric pathogens, and faecal shedding may contribute to the spread of these pathogens to the environment and the food chain [19, 20]. Improper implementation of food sanitation, failure to adhere to standard operational procedures and farm management techniques during the food production, poor handling and marketing of meat products can facilitate the transfer of diarrhoeal pathogens into food, water and its associated food products [21, 22]. Therefore, routine surveillance to determine the prevalence and emerging trends of AMR pathogens in cattle is important to ensure human health safety. The current study was designed to investigate the pathotype, virulence and resistance profiles of multidrug‐resistant (MDR) bacteria isolated from asymptomatic cattle faeces.
2. Materials and Methods
2.1. Ethics Approval and Sample Size
This study was conducted at the North‐West University (NWU), Mafikeng, South Africa. Ethics clearance for the study was obtained from the NWU‐AnimCare Committee, and ethics number NWU‐00771–23‐A5 was assigned to this study. The number of samples required for the study was determined using the following formula described by Charan and Biswas [23]:
where Z 1−α/2 = standard variate at 5% Type I error (p < 0.05) = 1.96, P = expected prevalence in population based on a previous study and d = absolute error or precision (which is 5%):
For the estimation of the prevalence of diarrhoeal pathogens, the sample size for this study was determined using the prevalence of 8% obtained by Belina et al. [24] to be the expected prevalence with a 95% confidence level and desired precision of 5% using the formula described by Charan and Biswas [23]. Accordingly, the minimum sample size required for the study was 246.
2.2. Sample Collection
Two hundred and sixty‐nine cattle faecal samples were obtained directly from the rectum of cattle with the help of veterinary personnel using sterile arm‐length gloves. Cattle with diarrhoea or clinical symptoms were not included. To avoid duplicate sampling, cattle were confined to their handling pens. The collected samples were promptly transferred on ice to the Molecular Microbiology Laboratory at NWU in a sterile sample collection box with the proper labelling for microbiological examination.
2.3. Isolation of Bacteria
For microbiological examination, only E. coli, V. cholerae and Salmonella species were targeted in this study due to their frequent causes of diarrhoea [6] and antibiotics contraindication [25]. Upon arrival in the laboratory, 1 g of each faecal sample was dissolved in 9 mL of 10% (w/v) saline solution. Aliquots of 100 μL from each of the tenfold serial dilutions were spread‐plated on eosin–methylene blue (EMB), Salmonella Shigella agar (SSA) and thiosulfate citrate bile sucrose (TCBS) agar, respectively. The plates were incubated aerobically at 37°C for 24 h.
Colonies with a green metallic sheen were selected from EMB agar for presumptive E. coli identification. Transparent/colourless colonies with or without a black centre were selected on SSA for presumptive Salmonella species. Yellowish colonies were selected from TCBS agar for presumptive identification of V. cholerae. All the selected presumptive bacterial isolates were purified by streaking on nutrient agar, and the plates were incubated aerobically at 37°C for 24 h. A single pure colony of each isolate was picked and cultured in nutrient broth at 37°C for 24 h. Thereafter, DNA extractions were performed from the overnight cultures using the Zymo Research Genomic DNA‐Tissue MiniPrep kit (Zymo Research Corp, Irvine, CA, USA) according to the manufacturer’s instructions.
2.4. Confirmation of the Isolated Bacteria Using PCR Assay
To confirm the identity of the bacterial isolates, PCR targeting species‐specific genes was performed. The PCRs comprised 12.5 μL PCR master mix (Dream Taq Green 2x, ThermoFisher, Waltham, MA, template DNA (1.0 μg/μL)), forward and reverse primers (0.25‐μL per reaction for each primer), and completed to a total volume of 25 µL with nuclease‐free water. Nuclease‐free distilled water was used as a negative control template. The PCR was performed in a thermal cycler (Bio‐Rad C1000 TouchTM Thermal Cycler) to amplify the targets under the conditions outlined in Table 1.
The PCR products were electrophoresed using the Mupid‐One Electrophoresis System (NIPPON Genetics, Tokyo, Japan) at 90 V for 45 min by using agarose gel (1.5%) stained with ethidium bromide at 0.5 μg/mL of the gel. Stained gels were examined with a horizontal piece of Pharmacia Biotech equipment (Model Hoefer HE99X, Amersham Pharmacia Biotech, Sweden) to visualize and capture gel images. The image was taken with Gene Snap (Version 6.00.22) software utilizing a ChemiDoc Imaging System (Bio‐Rad ChemiDocTM MP Imaging System, UK).
2.5. Antibiotic Susceptibility Test
The identified isolates (E. coli, V. cholerae and Salmonella spp.) were tested for antimicrobial susceptibility testing on Muller–Hinton agar by using the disc diffusion technique against a panel of eight antibiotics belonging to different categories of antimicrobial agents, as indicated in our previous study [10]. The pure colonies of E. coli, V. cholerae and Salmonella spp. were suspended and cultured overnight in nutrient broth at 37°C. Turbidity of each broth culture was checked against 0.5 McFarland standards. The organisms in the broth were aseptically and uniformly inoculated into Mueller–Hinton agar (Merck Co., Germany). The antibiotic discs were carefully applied to the surface of the inoculated agar and then incubated aerobically at 37°C for 24 h. The tests per isolate were performed in triplicate. The disc diffusion method was used for the susceptibility test in accordance with the Clinical and Laboratory Standards Institute guidelines. The antibiotics tested comprised azithromycin (15 μg), imipenem (10 μg), levofloxacin (5 μg), penicillin (10 μg), streptomycin (10 μg), sulfamethoxazole (23.75 μg), tetracycline (30 μg) and vancomycin (30 μg). The selected antibiotics are commonly used to treat diarrhoea in cattle [30]. After overnight incubation, the zones of inhibition were measured and interpreted as susceptible or resistant.
2.6. Biofilm Formation Assay
Biofilm formation was quantitatively assessed for all isolated bacteria. Briefly, 100 μL of the standardized isolate was transferred to 100 μL of double‐strength nutrient broth in a 96‐well microtiter plate and incubated at different temperatures (4°C and 25°C) for 24, 48 and 72 h. The microtiter well containing sterile distilled water and nutrient broth was used as a negative control. The planktonic cells and the spent medium were discarded from the microtiter plate after incubation. The microtiter plate was rinsed three times with sterile distilled water and then stained with 100 μL of 0.1% crystal violet (CV) solution for 1 h. The stained inside microtiter plate was then rinsed three times with sterile distilled water to remove the unbound dye. Thereafter, 95% ethanol was added to the microtiter plate with gentle mixing to release the bound CV dye from the biofilm, and 100 μL of the solution was transferred to a new sterile 96‐well microtiter plate for quantification at 600 nm on a plate reader. All the experiments were performed in triplicate.
2.7. Whole Genome Sequence Analysis
Three bacterial isolates that were previously confirmed phenotypically to be MDR and biofilm‐forming were selected for whole genome sequencing (WGS) [10] to get comprehensive genomic information. The three isolates included one E. coli (DEC‐NWU), one V. cholerae (DVC_NWU) and one Salmonella species (DSS_NWU). DNA was extracted from competent overnight cultures of DEC_NWU, DVC_NWU and DSS_NWU using the Zymo Research Genomic DNA‐Tissue MiniPrep kit (Zymo Research Corp, Irvine, CA, USA) according to the manufacturer’s instructions. Libraries were prepared using the MGI Universal DNA Library Prep Kit according to the manufacturer’s instructions. Paired‐end sequencing was done on the MGI DNBSEQ‐G400 for 312 cycles.
Raw reads were uploaded into the knowledgebase (KBase) platform Version 2.1 (https://kbase.us/). The sequence quality of the raw reads was verified using FastQC Version 0.11.5 (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/). The adaptors and ambiguous nucleotide sequences were filtered and trimmed using Trimmomatic Version 0.36 [31]. De novo assembly of the trimmed (good quality) reads was performed using SPAdes Version 3.13.0 [32, 33]. The genome assembly drafts were annotated using the prokaryotic genome annotation pipeline (PGAP) on the KBase platform (https://kbase.us/) and rapid annotation using the subsystem technology (RAST) server Version 2.0 for comparison [34, 35]. The annotated genome sequences were deposited into GenBank under the assigned accession numbers.
2.8. Detection of Pathotype and Phylogenetics of the Isolate
The group in which the isolates belong based on the pathogenicity was determined by searching the specific isolate contigs against the PathogenFinder Version 1.1 database (https://cge.food.dtu.dk/services/SerotypeFinder/) on the genomic epidemiology online platform [36] and as reported in a previous study [10]. To determine the genetic nexus of the DEC_NWU, DVC_NWU and DSS_NWU with other species, high‐quality genomes of E. coli, V. cholerae and S. enterica were selected from the National Centre for Biotechnology Information (NCBI) and included in the phylogenetic analysis. Mash/MinHash was used to find the nearest reference and representative genomes. To assess each genome’s phylogenetic position, PATRIC global protein families were chosen from these genomes. MUSCLE was used to align the protein sequences from these families, and the nucleotides for each sequence were then mapped to the protein alignment. A data matrix was created by concatenating the combined alignment sets for amino acids and nucleotides. Fast bootstrapping was used to estimate support values in the tree, and RaxML was used to evaluate the matrix.
2.9. Detection of Virulome and Resistome Harboured by the Isolates
The detection of the virulence genes harboured by the isolates was assessed by searching their contigs against the virulence factor database (VFDB) (Virulence Factors of Bacterial Pathogens (mgc.ac.cn)), as indicated in a previous study [10]. Also, the identification of acquired virulence genes was found through the VirulenceFinder Version 2.0 database (https://cge.food.dtu.dk/services/VirulenceFinder/). The AMR gene harboured by the isolates was predicted by uploading the contigs to the online resistance gene identifier (RGI) platform [37].
3. Results
3.1. Bacteria Detected
In this study, a total of 45 E. coli, 10 Salmonella species and 5 V. cholerae were obtained. Their characteristics, reported in a previous study [10], showed that they harboured species‐specific genes (Figures 1(a), 1(b), 1(c) and 1(d)).
FIGURE 1Agarose gel showing amplification for (a) bacteria, (b) Escherichia coli, (c) Salmonella species and (d) V. cholerae isolated from the faecal samples. Lane L is a 100‐bp DNA ladder, and Lane 1 is a negative control (PCR without DNA).(a)(b)(c)(d)
3.2. Phenotypic Antibiotic‐Resistant Profile of Isolated Bacteria
As indicated in a previous and related study [10], the antibiotic susceptibility test of the isolated bacteria revealed diverse resistance patterns against a panel of 8 antibiotics tested. The antibiotic susceptibility test result confirmed that E. coli, Salmonella species and V. cholerae isolates were resistant to azithromycin (100%, 100% and 100%), levofloxacin (0%, 20% and 20%), penicillin (100%, 100% and 100%), streptomycin (9%, 60% and 20%), sulfamethoxazole (49%, 90% and 40%), tetracycline (64%, 100% and 20%) and vancomycin (9%, 20% and 0%), respectively (Figure 2). Notably, all the isolated bacteria were susceptible to imipenem. A large proportion (73% E. coli, 100% Salmonella species and 80% V. cholerae) of the bacterial isolates were classified as MDR because they were resistant to at least one antibiotic in three or more antimicrobial categories.
Antibiotic resistance profile of the isolated bacteria (source [10]). The error bars represent the standard deviation.
3.3. Biofilm Formation of the Isolated Bacteria
Most of the bacteria (100% E. coli, 100% Salmonella species and 60% V. cholerae) isolated in this study formed biofilm at both temperatures (4°C and 25°C) and incubation periods (24, 48 and 72 h) tested. The isolates were classified as nonbiofilm formers (ODs ≤ ODc), weak biofilm formers (ODc < ODs < 2 × ODc), moderate biofilm formers (ODs = 2 × ODc) and strong biofilm formers (ODs > 2 × ODc). The cut‐off value (ODc) was defined as the optical density of the negative control (ODc), while ODs represent the optical density of the sample isolate. At 4°C, 59%, 76% and 52% of E. coli; 80%, 100% and 90% of Salmonella species; and 0%, 0% and 0% of V. cholerae isolates formed strong biofilms within incubation periods of 24, 48 and 72 h, respectively (Figure 3). Also, at 25°C, 100%, 100% and 100% of E. coli; 100%, 100% and 100% of Salmonella species; and 0%, 0% and 0% of V. cholerae isolates formed strong biofilms within incubation periods of 24, 48 and 72 h, respectively (Figure 4).
Biofilm formation of the isolated E. coli (EC), Salmonella species (SS) and V. cholerae (VC) at 4°C within 24, 48 and 72 h of incubation. The error bars indicate the standard deviation.
Biofilm formation of the isolated E. coli (EC), Salmonella species (SS) and V. cholerae (VC) at 25°C within 24, 48 and 72 h of incubation. The error bars indicate the standard deviation.
The contrast between V. cholerae and the consistently strong biofilm‐forming abilities of E. coli and Salmonella species highlights important interspecies differences in biofilm regulation and environmental adaptability. Based on a two‐way ANOVA, the formation of strong biofilms by E. coli and Salmonella species was significantly affected by temperature (p values = 0.5), whereas incubation time did not have a significant effect (p values = 0.034).
3.3.1. Whole Genome Sequence Analysis
3.3.1.1. Genome Assembly and Annotation
The three bacterial isolates (DEC_NWU, DCV_NWU and DSS_NWU) previously confirmed phenotypically to be MDR and biofilm‐forming sequenced revealed numerous genomic features. The PGAP on the KBase platform confirmed that the genome lengths of DEC_NWU, DCV_NWU and DSS_NWU were 4,803,571, 4,499,945 and 5,374,783 bp, respectively.
There were 99, 38 and 146 contigs with a guanine–cytosine (GC) content of 50.47%, 55.9% and 51.31% in the genomes of the DEC_NWU, DVC_NWU and DSS_NWU, respectively. The shortest sequence length at 50% (N50) in the genomes of the isolates DEC_NWU, DVC_NWU and DSS_NWU were 183,022, 659,577 and 246,095 bp, respectively, while the smallest number of contigs whose sum length produced N50 (L50) were 9, 2 and 7, respectively.
DEC_NWU, DVC_NWU and DSS_NWU were confirmed to belong to the super kingdom bacteria and possessed 4,758, 4,215 and 5624 protein‐coding sequences (CDS), respectively. Based on the Genome Taxonomy Database (GTDB) classification, DEC_NWU, DVC_NWU and DSS_NWU were confirmed to be E. coli, V. cholerae and S. enterica, respectively. The assembled genomes of DEC_NWU, DVC_NWU and DSS_NWU, deposited into the NCBI GenBank, were assigned accession numbers CP121294, CP122254 and CP123007, respectively. The annotated genome features and taxonomy of each isolate are summarized in Table 2.
3.3.1.2. Subsystem Analysis
The genomes of DEC_NWU, DVC_NWU and DSS_NWU contained subsystems, which are sets of proteins that together implement specific biological processes or structural complexes. The subsystem found in the genomes of DEC_NWU, DVC_NWU and DSS_NWU includes those responsible for metabolism, protein processing, stress response, defence mechanisms, membrane transport, energy, cell envelope, cellular processing, RNA and DNA processing. The genes responsible for the metabolism subsystem were predominant in the genomes of DEC_NWU, DVC_NWU and DSS_NWU. The gene frequencies of subsystems in the genomes of DEC_NWU, DVC_NWU and DSS_NWU are shown in Figures 5, 6 and 7, respectively.
Circos plot depicting the frequency of subsystems and their related genes contained in the genome of DEC_NWU isolate.
Circos plot depicting the frequency of subsystems and their related genes contained in the genome of DVC_NWU isolate.
Circos plot depicting the frequency of subsystems and their related genes contained in the genome of DSS_NWU isolate.
The ribbons in Figure 5 show the cell envelope (red = gene 4%), cellular processes (orange = gene 5%), defence and virulence (yellow = gene 8%), DNA processing (lemon = gene 4%), energy (screamin’ green = gene 13%), membrane transport (malachite = gene 7%), metabolism (downy = gene 38%), miscellaneous (viking = gene 5%), protein processing (blue = gene 11%) and RNA processing (purple = gene 3%) present in the genome of DEC_NWU isolate. While the ribbons in Figure 6 show the cell envelope (red = gene 4%), cellular processes (orange = gene 5%), DNA processing (lemon = gene 5%), defence virulence (yellow = gene 8%), energy (screamin’ green = gene 14%), membrane transport (malachite = gene 8%), metabolism (downy = gene 37%), miscellaneous (viking = gene 1%), protein processing (blue = gene 11%) and RNA processing (purple = gene 3%) present in the genome of DVC_NWU isolate. Also, the ribbons in Figure 7 show the cell envelope (red = gene 5%), cellular processes (orange = gene 8%), DNA processing (lemon = gene 5%), defence virulence (yellow = gene 8%), energy (screamin’ green = gene 13%), membrane transport (malachite = gene 7%), metabolism (downy = gene 39%), miscellaneous (viking = gene 4%), protein processing (blue = gene 9%) and RNA processing (purple = gene 3%) present in the genome of DSS_NWU isolate.
3.3.2. Pathotype and Phylogenetics of the Isolate
Isolate DEC_NWU, DVC_NWU and DSS_NWU were confirmed as human pathogens based on the PathogenFinder prediction. The genetic nexus of the isolates with other species confirmed that DEC_NWU, DVC_NWU and DSS_NWU were related to E. coli O104:H4 (88%), V. cholerae O1 (100%) and S. enterica serovar Typhimurium (100%), respectively (Figure 8).
Phylogenetic tree of DEC_NWU, DVC_NWU, DSS_NWU and the closest bacterial species. The red colour indicates our isolates (DEC‐NWU, DVC_NWU and DSS_NWU), and the black colour indicates reference bacterial species from NCBI. The values represent the percentage closeness. The tree scale is 0.03.
3.3.3. Virulence Gene Profile of the Isolate
The result from the virulence factors of bacterial pathogens online platform confirmed that DEC_NWU, DVC_NWU and DSS_NWU possessed several virulence genes in their genomes. The DSS_NWU genome possessed the highest number (65) of virulence genes compared to DVC_NWU (29) and DEC_NWU (40). The isolate DEC_NWU, DVC_NWU and DSS_NWU genomes harboured virulence genes encoding adherence, invasion, iron uptake, secretory system and toxin production (Table 3).
The number of adherence virulence genes detected in the genomes of DEC_NWU, DVC_NWU and DSS_NWU were 21, 8 and 25, respectively. Notably, pili and fimbriae adherence virulence factors were present in DEC_NWU and DSS_NWU but absent in the DVC_NWU genome. The three isolates (DEC_NWU, DVC_NWU and DSS_NWU) have three iron uptake virulence factors of different types. A greater number of invasion virulence factors were detected in the genomes of DEC_NWU (5) and DSS_NWU (5) compared to DVC_NWU (2). The number of secretion system virulence factors detected in the genomes of the DEC_NWU, DVC_NWU and DSS_NWU were 6, 2 and 6, respectively. Haemolysin E/cytolysin A (hlyE/clyA) toxin virulence gene was detected only in DEC_NWU and DVC_NWU, while cytolethal distending toxin B (cdtB) and pertussis‐like toxin A (ptlA) virulence genes were detected in DSS_NWU genome.
3.3.4. Biofilm Profile of the Isolate
Adherence genes are strongly attributed to biofilm formation and have been recognized as a passive virulence factor contributing to many infectious diseases. The biofilm/adherence genes detected in the DEC_NWU genome include E. coli common pilus (acpA, acpB, acpC, acpD), E. coli laminin‐binding fimbriae (elfA, elfC, elfD, elfG) and haemorrhagic E. coli pilus (hcpA, hcpB, hcpC). Also, lateral flagella (lafC, lafT, lfhA, nueA), flagellar motor switch protein (fliM) and flagellar biosynthesis protein (fliP) were detected in the DVC_NWU genome. While the DSS_NWU genome harboured afimbrial adhesin (afaB), curlin major subunit (csgA, csgB, csgC, csgD), Type I fimbriae (fimA, fimC, fimD, fimF) and Type IV pili (pilQ, pilR, pilS, pilW). The number of biofilm/adherence genes detected in the genomes of DEC_NWU, DVC_NWU and DSS_NWU were 17, 7 and 24, respectively, as indicated in Table 3.
3.3.5. Antibiotics‐Resistant Gene of the Isolate
The results from the RGI database confirmed that DEC_NWU, DVC_NWU and DSS_NWU each harboured several antibiotics‐resistant genes. The classes of drugs for which the DEC_NWU, DVC_NWU and DSS_NWU genomes harboured resistance genes included aminoglycosides, cephalosporins, peptides, macrolides, fluoroquinolones, glycopeptides, penems, lincosamides, tetracyclines and phenicols. More resistance genes were detected in the DEC_NWU genome (97) than in DVC_NWU (85) and DSS_NWU (89). A summary of the resistance genes detected in the genomes of DEC_NWU, DVC_NWU and DSS_NWU, along with the corresponding drug class, is provided in Table 4.
4. Discussion
Several enteric bacteria, including Campylobacter jejuni, Escherichia coli, Vibrio cholerae, Salmonella and Shigella species, are among the notable causative agents of diarrhoea [6, 13, 14]. Their pathogenicity is enhanced by their ability to express a variety of virulence factors associated with colonization, invasion, toxin production, adhesion, iron acquisition, production of lipopolysaccharide capsules and secretion systems [15]. Contamination of the food chain, coupled with the consumption of undercooked food products, accounts for several outbreaks of diarrhoea [16, 17], and thus, ruminants, especially cattle, are considered the principal reservoir for these pathogens [19, 20]. The current study was designed to investigate the pathotype, virulence and resistance profiles of MDR bacteria isolated from asymptomatic cattle faeces.
In this study, a total of 45 E. coli, 10 Salmonella species and 5 V. cholerae were isolated from cattle faecal samples collected from different farms in the North‐West Province, South Africa. The phenotypic antibiotic susceptibility results revealed that E. coli, Salmonella spp. and V. cholerae isolates exhibited high resistance to azithromycin (100%), penicillin (100%) and tetracycline (64%–100%), with moderate to high resistance against sulfamethoxazole and streptomycin. Notably, all isolates were susceptible to imipenem.
Isolates DEC_NWU, DVC_NWU and DSS_NWU were phenotypically MDR and strong biofilm formers. The whole genome sequence analysis confirmed that DEC_NWU, DVC_NWU and DSS_NWU had genome lengths of 4,803,571 bp, 4,499,945 bp and 5,374,783 bp, respectively. Comparisons of the DEC_NWU, DVC_NWU and DSS_NWU genomes with other genomes within the same species in the PATRIC and NCBI databases confirmed that the isolates were E. coli, V. cholerae and S. enterica, respectively. In addition, PathogenFinder classified the isolates as human pathogens based on the virulence markers present in their genomes, when compared with genomes available in the PathogenFinder database [36]. Several studies have reported the presence of human pathogens in cattle faeces [20, 38]. Manishimwe et al. [39] detected the presence of pathogenic Salmonella species and E. coli from cattle faecal samples. Montso et al. [40] reported the presence of pathogenic E. coli in cattle faecal samples collected at a farm in the North‐West Province. Surveillance of the prevalence of pathogenic bacteria in cattle faecal samples cannot be neglected because of the morbidity and/or mortality it might cause if it contaminates food products [41]. Pathogenic bacteria in cattle faeces can contaminate the area used for slaughter, the cutting machines and workers during cattle processing for food. These pathogens can be transferred to food, food products and water if proper hygiene is not applied, especially at the processing and sales points.
The genetic nexus with other species confirmed that DEC_NWU, DVC_NWU and DSS_NWU were related to E. coli O104:H4 (88%), V. cholerae O1 (100%) and S. enterica serovar Typhimurium (100%), respectively. Outbreaks involving Vibrio, Escherichia coli and Salmonella represent a persistent and multifaceted threat to public health. For instance, E. coli O104:H4 was responsible for the large‐scale outbreak in Germany in 2011, which caused more than 830 cases of haemolytic uremic syndrome and 46 deaths [42, 43]. Likewise, Vibrio cholerae is a major global pathogen, endemic in many regions and responsible for large numbers of cholera cases every year [44–46]. In parallel, S. enterica serovar Typhimurium remains among the most frequently implicated agents in gastrointestinal disease outbreaks [47, 48]. These outbreak dynamics point to critical needs for enhanced surveillance systems, stricter food‐handling and cooking practices (especially for high‐risk foods) and targeted public health interventions to reduce morbidity, mortality and economic burden.
Data generated from bacterial pathogen VFDBs confirmed that the DEC_NWU, DVC_NWU and DSS_NWU genomes possess a wide range of virulence genes. The virulence factors harboured by the three isolated genomes include adherence, invasion, iron uptake, secretory system and toxin. The DEC_NWU genomes harboured adherence virulence gene encoding haemorrhagic E. coli pilus (hcpA, hcpB, hcpC), CFA/I fimbriae (cfaA, cfaB, cfaC, cfaD, cfaE), curli fibres (cgsA, cgsB, cgsC, cgsD, cgsE, cgsF, cgsG), E. coli common pilus (acpA, acpB, acpC, acpD, acpE, acpR), Type I fimbriae (fimA, fimB, fimC, fimD, fimE, fimF, fimG, fimH, fimI). The adherence gene in the DVC_NWU genome includes lateral flagella (lafC, lafT, lfhA, nueA). DSS_NWU genome possessed adherence genes that include a fimbrial adhesin (afaB), curlin major subunit (csgA, csgB, csgC, csgD, csgE, csgF, csgG), Type I fimbriae (fimA, fimC, fimD, fimF, fimH, fimI, fimW, fimY, fimZ) and Type IV pili (pilQ, pilR, pilS, pilW). Several classes of extracellular organelles in bacteria, which facilitate attachment and motility, have been associated with virulence factors that lead to pathogenicity [49]. These extracellular organelles include flagella, pili and curli fibres, which can be classified as adherence virulence factors [50].
The virulome composition of these isolates, including adherence, iron uptake and secretion systems, supports their potential to manipulate host immunity [51]. Mechanisms may similarly operate in biofilm‐forming pathogenic bacteria. Furthermore, the detection of multiple secretion systems and invasion‐associated factors suggests that the transforming growth factor‐beta signalling axis may be relevant in shaping host–pathogen interactions [52]. Similar pathways could be exploited by E. coli, V. cholerae and S. enterica to modulate antimicrobial responses during infection. In addition, environmental signals encountered by pathogens within the host or environment may trigger downstream regulatory networks that influence resistance and virulence gene expression [53]. Comparable mechanisms may be involved in regulating the gene expression profiles observed in the MDR, biofilm‐forming isolates described in this study.
VFDB analyses revealed that DEC_NWU, DVC_NWU and DSS_NWU genomes harboured toxin virulence genes encoding haemolysin/cytolysin A (hlyE/clyA), heat‐stable cytotonic enterotoxin (ast) and typhoid toxin (cdtB, pltA, pltB), respectively. Bacteria produce toxins as a long‐term survival strategy in the environment [54]. The toxin’s mode of action has been associated with virulence factors, causing pathological damage to susceptible hosts [55, 56]. Bacterial toxins can damage the cell membrane, inhibit protein synthesis, destroy blood cells and activate the immune response, leading to disease [56]. For example, haemolysin/cytolysin A (hlyE/clyA) has been reported to lyse the erythrocyte and mammalian cells of a susceptible host to form a pore [57]. Also, Dubreuil [58] reported that heat‐stable cytotonic enterotoxin (ast) induced intestinal fluid accumulation, leading to diarrhoeal illness.
Several genes that mediate resistance to 10 different drug classes were identified in the DEC_NWU, DVC_NWU and DSS_NWU genomes. The DEC_NWU, DVC_NWU and DSS_NWU genomes contained genes providing resistance to aminoglycosides, cephalosporins, peptides, macrolides, glycopeptides, fluoroquinolones, penems, lincosamides, tetracyclines and phenicols. Notably, the DEC_NWU genome harboured more resistance genes (97) than the DSS_NWU (89) and DVC_NWU (86) genomes. In addition, genes conferring resistance to aminoglycoside (mtdA, mtdb, mtdC, baeS, cpxA, baeR), tetracycline (acrA, sdiA, acrAB-TolC, marR, soxR, kpnF, kpnE), fluoroquinolone (mtdH, emrA, emrB, emrR, acrA, acrB acrE, acrF, acrS, rsmA, mdtM, soxS, H‐NS*, mdtE, gadX,* CRP*, evgA*), erythromycin (ermB), phenicol (acrS, acrA, acrB rsmA, mdtM, marA kpnE, kpnF) and macrolide (kpnF, kpnE, H‐NS, CRP) were relatively common in the genomes of the three isolates.
Efflux pump–associated genes such as acrA, acrAB-TolC, marA/marR, soxR/soxS, rsmA, H‐NS*, kpnE* and kpnF were widely distributed across the isolates. This pattern is consistent with previous studies, which have shown that MDR in Enterobacterales is often driven by global regulatory systems and efflux mechanisms, rather than by class‐specific resistance genes alone [59]. DEC_NWU exhibited the highest diversity of AMR determinants, including genes linked to aminocoumarin, fluoroquinolone, phenicol, rifamycin and tetracycline resistance, suggesting a strongly adaptive genomic background. DVC_NWU and DSS_NWU shared several resistance markers, such as aac(6′)-Iy, acrAB-TolC, soxR, H-NS, arnT, pmrF and bacA, reflecting the conserved nature of lipid A modification pathways and efflux‐associated resistance reported in environmental and clinical isolates [60]. The presence of vanG, albeit alone, across all isolates indicates a baseline glycopeptide tolerance, as previously described for van gene clusters in bacteria [61]. Furthermore, the detection of phosphonic acid–associated genes (uhpT, glpT) in DVC_NWU and DSS_NWU aligns with known fosfomycin‐resistance mechanisms involving transporter gene mutations or loss [62]. The widespread distribution of multidrug efflux and regulatory genes indicates substantial cross‐resistance capacity, raising significant public health and food safety concerns.
The genotypic data strongly support the observed phenotypic resistance patterns. For example, macrolide resistance (azithromycin) corresponds with the detection of efflux pump–associated genes (kpnE, kpnF, emrE, H‐NS, crp, evgA, and gadX) across all isolates, which are known to confer reduced intracellular drug accumulation [63]. Similarly, the universal penicillin resistance observed aligns with the presence of multidrug efflux components (acrA, acrE, acrF, acrS, marA, soxS, H‐NS), as well as global transcriptional regulators (marA, marR, soxR, and soxS) that upregulate β‐lactam resistance mechanisms [64]. In DEC_NWU and DSS_NWU Salmonella and E. coli, high resistance rates to tetracycline correlate with the presence of tetracycline‐associated efflux genes (acrA, acrB, mdtM, emrY, emrK, and mdfA), along with their regulators (marA, soxS, and H‐NS). Sulfamethoxazole resistance observed in the isolates can be attributed to efflux systems and global stress response regulators, which, although not sul‐specific genes (sul1 and sul2), can still mediate trimethoprim–sulfamethoxazole resistance [65].
The resistance to streptomycin observed in the isolates can be explained by aminoglycoside resistance determinants such as aac(6′)-Iy and multidrug efflux systems (kdpE, baeR, baeS, kpnE, kpnF) [66]. Although genes associated with fluoroquinolone resistance were present in all isolates, DEC_NWU did not exhibit phenotypic levofloxacin resistance. Also, the vancomycin resistance gene (vanG) was present in all three species, but DVC_NWU did not exhibit vancomycin resistance phenotypically. Similarly, all isolates were phenotypically susceptible to imipenem, despite harbouring genes such as marA, acrA and sdiA in their resistomes. This finding aligns with previous reports, indicating that the mere presence of resistance genes does not necessarily result in phenotypic resistance, as factors such as gene expression levels, regulatory activation and efflux pump dynamics can modulate susceptibility [67].
The AMR development in bacteria of animal origin is a risk to human health because of horizontal gene transfer via genetic mobile elements, conferring resistance to novel pathogenic organisms [68]. The presence of several AMR genes in the genomes of the three isolates poses a threat to public health, as there are limited antibiotics available to treat infections caused by these bacteria. AMR among pathogenic bacteria has been established as a public health problem due to the detrimental burden placed on patients and the economy [69]. Diseases caused by the AMR organisms can result in serious illnesses, prolonged hospital admissions, an increase in healthcare costs, a burden on society’s resources and treatment failures [70]. The increasing incidence of AMR emergence and spread among pathogens is a global health concern that requires urgent multisectoral action.
The virulome and resistome profiles of the isolates provide important insights into their potential linkage to known high‐risk and outbreak‐associated bacterial lineages. The DEC_NWU isolate’s combination of multidrug efflux systems and adherence factors suggests it may belong to a diarrheagenic E. coli lineage, such as enterotoxigenic E. coli (ETEC), although confirmation of pandemic clones, such as ST131, requires further typing. The DVC_NWU isolate’s possession of lateral flagella genes and macrolide resistance‐associated efflux pumps supports its possible affiliation with epidemic Vibrio cholerae El Tor lineages, pending verification of key toxin genes. Finally, the DSS_NWU isolate exhibits a comprehensive set of virulence determinants and MDR mechanisms characteristic of epidemic Salmonella serovars, including possible invasive or typhoidal strains, which may represent high‐risk clones such as ST11.
5. Limitations and Future Directions
This study was limited by its geographic sampling, as all cattle faecal samples were collected from a single region, which may not fully represent the genetic diversity and AMR profiles of bacterial pathogens in cattle from other regions or production systems. Additionally, only three isolates (DEC_NWU, DVC_NWU and DSS_NWU) were subjected to WGS, limiting the breadth of insights into the resistome and virulome across the broader bacterial community in cattle faeces. The cross‐sectional study design also precludes assessment of temporal variation in pathogen carriage and resistance patterns. Furthermore, environmental and management factors that may influence the prevalence and transmission of MDR pathogens were not evaluated. These findings provide valuable insights and serve as the basis for future studies.
Future studies should aim to include larger, geographically diverse samples to capture regional variation in AMR profiles. Longitudinal surveillance could provide insight into seasonal or management‐related trends in pathogen shedding. Expanding the number of isolates subjected to WGS and integrating metagenomic approaches may reveal broader patterns of the resistome and virulome within the cattle gut microbiome. Moreover, incorporating environmental sampling (water, soil and feed) along with farm management data could help identify critical control points to reduce the dissemination of MDR, biofilm‐forming pathogens into the food chain.
6. Conclusion
The current findings revealed that cattle harbour biofilm‐forming, MDR bacteria asymptomatically. WGS revealed that some of these bacteria were human pathogens. Several virulence and AMR genes were detected in the genomes of the sequence isolates (DEC_NWU, DVC_NWU and DSS_NWU) associated with known high‐risk and outbreak‐related bacterial lineages. Thus, continuous monitoring of pathogenic bacteria and enforcement of control measures to minimize contamination in the farm environment and in abattoirs are imperative to safeguard human health and ensure food safety. Further research is needed to develop novel alternative on‐farm strategies to combat the dissemination of MDR pathogenic bacteria that pose a threat to the safety and security of our food supply.
NomenclatureAMRAntimicrobial resistanceCDSCoding sequencesDNADeoxyribonucleic acidPGAPProkaryotic genome annotation pipelineRASTRapid annotation using the subsystem technologyRGIResistance gene identifierrRNARibosomal ribonucleic acidtRNATransfer ribonucleic acidWGSWhole genome sequencing
Author Contributions
C.N.A., A.O.A., O.E.F. and T.O.A.: conceptualization and resources. T.O.A., D.J.A., B.O.O., P.K.M., A.J.E., J.E‐N., T.R.C., O.E.F., A.O.A. and C.N.A.: experimental design, analysis, interpretation and writing. C.N.A. and T.O.A.: experimenting. C.N.A., A.O.A. and O.E.F.: supervision.
Funding
This study was supported by North‐West University, no Grant number.
Disclosure
All authors contributed to the article and approved the submitted version.
Consent
All authors consent to its submission for publication.
Conflicts of Interest
The authors declare no conflicts of interest.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Pires S. M. and Devleesschauwer B. , Morris J. G. and Vugia D. J. , Chapter 1-Estimates of Global Disease Burden Associated With Foodborne Pathogens, Foodborne Infections and Intoxications, 2021, 5th edition, Academic Press, 3–17, 10.1016/B 978-0-12-819519-2.00020-7. · doi ↗
- 2Mullaney S. B. , Hyatt D. R. , Salman M. D. , Rao S. , and Mc Cluskey B. J. , Estimate of the Annual Burden of Foodborne Illness in Nondeployed Active Duty US Army Service Members: Five Major Pathogens, 2010–2015, Epidemiology and Infection. (2019) 147, 10.1017/S 0950268818003199, 2-s 2.0-85065644317.31063088 · doi ↗ · pubmed ↗
- 3Lakhan C. , Badrie N. , Ramsubhag A. , and Indar L. , Detection of Foodborne Pathogens in Acute Gastroenteritis Patient Stool Samples Using the ‘Biofire’ Filmarray Gastrointestinal PCR Panel in the Republic of Trinidad and Tobago, West Indies, Microorganisms. (2022) 10, no. 8, 10.3390/microorganisms 10081601.PMC 941617636014019 · doi ↗ · pubmed ↗
- 4Davedow T. , Carleton H. , Kubota K. et al., Pulsenet International Survey on the Implementation of Whole Genome Sequencing in Low and middle-income Countries for Foodborne Disease Surveillance, Foodborne Pathogens and Disease. (2022) 19, no. 5, 332–340, 10.1089/fpd.2021.0110.35325576 PMC 10863729 · doi ↗ · pubmed ↗
- 5Vos T. , Lim S. S. , Abbafati C. et al., Global Burden of 369 Diseases and Injuries in 204 Countries and Territories, 1990–2019: a Systematic Analysis for the Global Burden of Disease Study 2019, The Lancet. (2020) 396, no. 10258, 1204–1222, 10.1016/S 0140-6736(20)30925-9.PMC 756702633069326 · doi ↗ · pubmed ↗
- 6Behera D. K. and Mishra S. , The Burden of Diarrhea, Etiologies, and Risk Factors in India from 1990 to 2019: Evidence from the Global Burden of Disease Study, BMC Public Health. (2022) 22, no. 1, 10.1186/s 12889-022-12515-3.PMC 875919635027031 · doi ↗ · pubmed ↗
- 7Wang H. , Naghavi M. , Allen C. et al., Global, Regional, and National Life Expectancy, all-cause Mortality, and Cause-specific Mortality for 249 Causes of Death, 1980–2015: a Systematic Analysis for the Global Burden of Disease Study 2015, Lancet Global Health. (2016) 388, no. 10053, 1459–1544, 10.1016/S 0140-6736(16)31012-1, 2-s 2.0-84994158650.PMC 538890327733281 · doi ↗ · pubmed ↗
- 8Murray C. J. L. , Aravkin A. Y. , Zheng P. et al., Global Burden of 87 Risk Factors in 204 Countries and Territories, 1990–2019: A Systematic Analysis for the Global Burden of Disease Study 2019, The Lancet. (2020) 396, no. 10258, 1223–1249, 10.1016/S 0140-6736(20)30752-2.PMC 756619433069327 · doi ↗ · pubmed ↗