Antimicrobial resistance and virulence gene profiles of Enterococcus faecalis and Enterococcus faecium isolated from subclinical bovine mastitis milk and cow dung Authors: Ntelekwane George Khasapane, Kgaugelo Lekota, Mofokeng Lehlohonolo, Jane Nkhebenyane, Oriel Thekisoe, Tsepo Ramatla Published: 2025 ABSTRACT Subclinical mastitis poses a hidden threat to dairy productivity and animal health, often harbouring antimicrobial-resistant pathogens. It is becoming increasingly recognized that Enterococcus species cause mastitis in dairy cows. Accurately characterizing the regional epidemiology of enterococcal mastitis, determining its correlations with management variables, and comprehending its effects on udder health all depend on accurate species information. This study investigated the occurrence, antibiotic resistance and virulence factors of Enterococcus faecalis and Enterococcus faecium in cow dung and milk samples from cows with subclinical mastitis. Subclinical mastitis was identified in 39.0% (68/174) of cows and 27.8% (194/696) of quarters, based on results from the California Mastitis Test (CMT) and somatic cell counts (SCC), respectively. Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF-MS) and Polymerase Chain Reaction (PCR) targeting the ddl gene confirmed the predominance of E. faecalis (93%) and E. faecium (6.4%) in milk samples, while cow dung samples yielded only E. faecalis (100%). Notably, among the E. faecalis isolates from milk samples, 17.2% exhibited vancomycin resistance, whereas streptomycin resistance was found in a smaller proportion of isolates (6.8%). All (100%) E. faecium isolates from the same milk samples showed resistance to vancomycin. The findings also revealed that 11 (32.3%) of E. faecium isolates from cow dung were resistant to vancomycin. Multidrug resistance (MDR) was observed in 20.6% of milk and 6.8% of cow dung isolates. The vanA gene was the most prevalent antibiotic resistance gene (ARG), detected in 96% of E. faecalis isolates. Virulence profiling of Enterococcus spp. isolates showed varying gene prevalence in milk (asa1: 33.3%, ace: 12.7%, esp: 10%) and cow dung samples (gelE: 53.2%, hyl: 38.2%). This study has indicated a significant occurrence of antimicrobial-resistant E. faecalis and E. faecium strains obtained from subclinical cattle mastitis. These findings emphasize the role of Enterococcus spp., especially vancomycin-resistant strains, as emerging threats in bovine subclinical mastitis, with possible implications for zoonotic transmission and antimicrobial stewardship in dairy systems. Introduction Bovine mastitis is a major infectious disease of dairy cattle, characterized by physicochemical alterations in milk that compromise its composition and quality. It is considered an endemic disease with significant economic consequences for the global dairy industry1. Although antibiotics remain the primary therapeutic option, treatment efficacy is often undermined by the emergence of antimicrobial resistance (AMR), resulting in reduced cure rates and the persistence of opportunistic bacterial infections, including those caused by the Enterococcus species2,3. Within this genus, E. faecalis and E. faecium are the most clinically relevant, together accounting for more than 80% of enterococcal infections worldwide. They are recognized as important opportunistic pathogens and are ranked as the third and fourth most common nosocomial agents globally4,5. Enterococcus species are ubiquitous in nature, capable of contaminating the food chain, colonizing the intestinal tract of both humans and animals, and serving as indicators of faecal contamination in water6,7. The predominant species identified in both human and animal microbiota include E. faecalis, E. faecium, E. hirae, and E. durans8. In dairy cattle, Enterococcus can colonize the teat skin and contribute to the pathogenesis of mastitis9,10. Therapeutic management of enterococcal infections is complicated by the high level of resistance to commonly used antibiotics, including bacteremia, endocarditis, urinary tract infections, and pelvic infections11. Antibiotic resistance is a natural aspect of bacterial evolution; however, using these compounds in veterinary medicine has accelerated this process by promoting the growth of microorganisms capable of surviving adverse conditions12. The bacterial species comprising the ESKAPE group, Enterococcus species, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacteriaceae, are involved explicitly in resistance phenomena13. Furthermore, enterococcal strains are becoming resistant to last-resort antibiotics, including daptomycin and linezolid14,15. The pathophysiology of E. faecalis is influenced by virulence factors and antibiotic resistance. According to Chajęcka-Wierzchowska et al.16, numerous studies have identified virulence factors in E. faecalis that may contribute to the severity of infections in humans and animals. Research has documented the existence of virulence factors (agg and ace) that promote E. faecalis adhesion and colonization as well as cytolysis and host dissemination (cyl, gelE, and sprE)17. Additionally, E. faecalis possesses biofilm-forming machinery (pili, gelE, and fsr quorum-sensing systems) that provides additional antimicrobial resistance and enables the bacterium to cling to biotic and abiotic surfaces17. Milk-borne pathogen surveillance in South Africa suffers from limited geographic coverage, reliance on cross-sectional and composite sampling, a lack of longitudinal data, and underrepresentation of smallholder dairy systems. As a result, regions like the Free State’s Thabo Mofutsanyana district remain under-studied. The current study aimed to detect and identify Enterococcus spp. from subclinical mastitis cow milk and their associated cow dung samples, including virulence and antibiotic resistance profiles. Materials and methods Ethical consideration All procedures and methods were carried out following relevant guidelines and regulations and were reported in accordance with ARRIVE guidelines (https://arriveguidelines.org). No anaesthesia or euthanasia methods were employed or involved in the present study. The study was further authorized by the University of the Free State’s Animal Research Ethics Committee (UFS-AED2024/0015) and the Department of Agriculture, Land Reform and Rural Development by section 20 of the Animal Diseases Act, 1984 (Ref Number: 12/11/1/4/6 (6335 ON)) and all experimental protocols were approved by the two named institutional and/or licensing committee . Study design, sampling strategy and initial screening of SCM In this cross-sectional study, the authors collected 174 milk samples from dairy cows across six farms in the Thabo Mofutsanyana District Municipality, Free State Province (Fig. 1). A total of 174 composite milk samples were collected using 50 mL sterile containers from four local municipalities: Mantsopa (40), Bohlokong (40), Maluti-a-Phofung (40), and Setsoto (54). Following the National Mastitis Council’s (NMC) guidelines, aseptic milk samples were obtained from cows that appeared to be healthy, from each quarter. Fig. 1Geographical locations of sample collection sites within the Free State Province, South Africa [(https://www.mapchart.net/world.html (accessed on 02 May 2025)]. Geographical locations of sample collection sites within the Free State Province, South Africa [(https://www.mapchart.net/world.html (accessed on 02 May 2025)]. A cow was screened for subclinical mastitis on the farm using the California Mastitis Test (CMT). Before sampling, each cow’s udder was cleaned with distilled water and dried with a disposable paper towel to prevent cross-contamination. After cleaning the udder and disinfecting the teats with 70% ethanol, 10 ml of milk was drawn from each cow. Next, 2 mm of milk samples from each lactating quarter that were free of clinical mastitis were milked and put into different cups on the CMT paddle. By tilting the CMT paddle and decanting the extra milk, the milk samples were equalized. Each cup of the CMT paddle containing the milk received an equal amount of CMT reagent, which was well mixed for 15 s. Thereafter, positive cow samples were then checked for SCM using the somatic cell count (SCC) assay (Mérieux NutriSciences, South Africa) and using Karzis et al.18 recommended scoring and interpreting of the level of inflammation based on the CMT and SCC results were scored as follows: 0 (negative (healthy quarter), somatic cell count (SCC) ≤ 100,000 cells/mL milk), 1+ (weak positive, SCC > 100,000 < 500,000 cells/mL milk), 2+ (distinct positive, SCC > 500,000 < 1000,000 cells/mL milk), and 3+ (strong positive, SCC ≥ 1000,000 cells/mL milk). Furthermore, faecal samples were collected directly from the rectum of each animal deemed to have subclinical mastitis and placed in a 50 mL sterile container. All samples were processed within 12 h of collection to ensure optimal preservation and integrity of the specimens19. Before the analysis, the samples were thawed in a refrigerator. Isolation and identification of Enterococcus species After plating 100 µL of milk onto Slanetz and Bartley Agar (SBA) (Oxoid, UK), the samples were incubated at 37 °C for 24 to 48 h. The sample culture was considered negative if no bacteria could be isolated after the direct plating procedure. A total of 68 cow dung samples from subclinically ill cows (10 g) were homogenized in 90 mL of a 2% sterile Buffered Peptone water (Oxoid, UK) for 2 min in a Stomacher BagMixer (Interscience, St. Nom, France). Samples were manually subjected to serial dilutions up to 10–8 dilution factor and then plated on Slanetz and Bartley agar (Oxoid, UK), followed by incubation at 37 °C for 24–48 h. The selection of enterococcal strains was aided by developing dark reddish colonies on the surface of the utilized media following cultivation. Following double purification, pure cultures from distinct colonies were identified as potential Enterococcus species. The isolates were then identified using matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry with MALDI-Biotyper 3.0 software (Bruker Daltonics, Bremen, Germany), as previously reported by Cameron et al.20. Two identification attempts were performed for each isolate. Isolates were considered unidentified if they could not be identified after two rounds of MALDI-TOF MS analysis. A cut-off score of ≥ 1.7 was used as the threshold for quality control. For further research, all pure Enterococcus spp. isolates were stored at -80 °C in Brain Heart Infusion (BHI) broth with 15% glycerol. Genomic DNA extraction of Enterococcus species A colony grown overnight at 37 °C on nutrient agar plates had its DNA extracted. Each colony was vortexed for two minutes after being inoculated into 200 µL of sterile distilled water. Thermo Fisher Scientific, Germany, centrifuged the cells for ten minutes at 13,000 rpm. The cells were lysed for 15 min at 100 °C in a heat block following the pipetting of 500 µL of distilled water into the Eppendorf tubes and vortexing21. The cell debris was separated using a Mini Spin centrifuge (Thermo Fisher Scientific, USA), which centrifuged the same sample for five minutes at 10,000 rpm. The supernatant was used immediately as a PCR template after extraction. Molecular identification of the isolates Species identification of Enterococci was further confirmed using singleplex-PCR targeting the species-specific ddlE gene. The primers employed were specific to E. faecium (550 bp amplicon; F: TAGAGACATTGAATATGCC, R: TCGAATGTGCTACAATC) and E. faecalis (941 bp amplicon; F: ATCAAGTACAGTTAGTCT, R: ACGATTCAAAGCTAACTG), as previously described by Jannati et al.22. The Enterococcus spp. DNA template (1 µL), 10 µL of Taq 2X Master Mix from New England Biolabs, which is distributed by Inqaba Biotech (South Africa), 1 µM per primer (1 µL), and 7 µL of RNase water made up a total of 20 µL PCR reaction. The PCR amplifications were conducted following the established method using a thermal cycler (Eppendorf® Mastercycler Personal). The cycling parameters for PCR consisted of an initial denaturation step at 94 °C for 2 min, followed by 30 cycles of denaturation at 94 °C for 1 min, annealing at 54 °C for 1 min, extension at 72 °C for 1 min, and a final extension step at 72 °C for 10 min. Both E. faecalis ATCC 29212 and E. faecium ATCC 700221 strains were utilized as positive controls. The PCR amplicons (5 µL) were analysed in a 1.2% agarose gel electrophoresis using ethidium bromide staining (0.5 µg/mL) and at 100 volts for 30 min. PCR analysis was used to identify Enterococcus spp., with E. faecalis species yielding a 941 bp product and E. faecium species yielding a 550 bp product. Moreover, due to budgetary constraints, the PCR products were not sent for sequencing to obtain their accession numbers. Molecular detection of Enterococcus species using virulence genes Six virulence genes were screened among the Enterococcus isolates (Table 1). These included genes associated with aggregation substances (asa1), enterococcal surface protein (esp), gelatinase (gelE), glycosyl hydrolase (hly), cytolysin (cylA), and genes encoding collagen-binding protein (ace). Table 1Primer sequences of antimicrobial-resistant and virulent genes screened among Enterococcus species.GenesPrimer sequence (5´-3´)Annealingtemp (°C)Ampliconsize (bp)ReferencesAMR genes tetK TCGATAGGAACAGCAGTA55 °C844Bag et al.23CAGCAGATCCTACTCCTT tetL TCGTTAGCGTGCTGTCATTC55 °C267GTATCCCACCAATGTAGCCG tetM GTGGACAAAGGTACAACGAG55 °C406CGGTAAAGTTCGTCACACAC tetS CATAGACAAGCCGTTGACC55 °C667ATGTTTTTGGAACGCCAGAG int GCGTGATTGTATCTCACT50 °C1028Morandi et al.24GACGCTCCTGTTGCTTCT ermB CATTTAACGACGAAACTGGC55 °C738Jensen et al.25GGAACATCTGTGGTATGGCG vatD GCTCAATAGGACCAGGTGTA58 °C361–632Jensen et al.22TCCAGCTAACATGTATGGCG vatE ACTATACCTGACGCAAATGC58 °C361–632GGTTCAAATCTTGGTCCG vanA TCTGCAATAGAGATAGCCGC55 °C377Jensen et al.22GGAGTAGCTATCCCAGCATT vanB GCTCCGCAGCCTGCATGGACA58 °C529ACGATGCCGCCATCCTCCTGCVirulence genes asa1 GCACGCTATTACGAACTATGA56 °C375Morandi et al.24TAAGAAAGAACATCACCACGA esp AGATTTCATCTTTGATTCTTGG56 °C510AATTGATTCTTTAGCATCTGG gelE TATGACAATGCTTTTTGGGAT56 °C213AGATGCACCCGAAATAATATA hyl ACAGAAGAGCTGCAGGAAATG56 °C276GACTGACGTCCAAGTTTCCAA cylA ACTCGGGGATTGATAGGC55 °C688GCTGCTAAAGCTGCGCTT ace GAATTGAGCAAAAGTTCAATCG55 °C1008GTCTGTCTTTTCACTTGTTTC Table 1Primer sequences of antimicrobial-resistant and virulent genes screened among Enterococcus species.GenesPrimer sequence (5´-3´)Annealingtemp (°C)Ampliconsize (bp)ReferencesAMR genes tetK TCGATAGGAACAGCAGTA55 °C844Bag et al.23CAGCAGATCCTACTCCTT tetL TCGTTAGCGTGCTGTCATTC55 °C267GTATCCCACCAATGTAGCCG tetM GTGGACAAAGGTACAACGAG55 °C406CGGTAAAGTTCGTCACACAC tetS CATAGACAAGCCGTTGACC55 °C667ATGTTTTTGGAACGCCAGAG int GCGTGATTGTATCTCACT50 °C1028Morandi et al.24GACGCTCCTGTTGCTTCT ermB CATTTAACGACGAAACTGGC55 °C738Jensen et al.25GGAACATCTGTGGTATGGCG vatD GCTCAATAGGACCAGGTGTA58 °C361–632Jensen et al.22TCCAGCTAACATGTATGGCG vatE ACTATACCTGACGCAAATGC58 °C361–632GGTTCAAATCTTGGTCCG vanA TCTGCAATAGAGATAGCCGC55 °C377Jensen et al.22GGAGTAGCTATCCCAGCATT vanB GCTCCGCAGCCTGCATGGACA58 °C529ACGATGCCGCCATCCTCCTGCVirulence genes asa1 GCACGCTATTACGAACTATGA56 °C375Morandi et al.24TAAGAAAGAACATCACCACGA esp AGATTTCATCTTTGATTCTTGG56 °C510AATTGATTCTTTAGCATCTGG gelE TATGACAATGCTTTTTGGGAT56 °C213AGATGCACCCGAAATAATATA hyl ACAGAAGAGCTGCAGGAAATG56 °C276GACTGACGTCCAAGTTTCCAA cylA ACTCGGGGATTGATAGGC55 °C688GCTGCTAAAGCTGCGCTT ace GAATTGAGCAAAAGTTCAATCG55 °C1008GTCTGTCTTTTCACTTGTTTC Primer sequences of antimicrobial-resistant and virulent genes screened among Enterococcus species. Antimicrobial susceptibility testing Antimicrobial susceptibility testing was performed using the disc diffusion method on Mueller-Hinton agar (Merck, Germany). The discs were put on a plate and incubated aerobically for 18 to 24 h at 37 °C following spreading the overnight inoculum containing the isolates. The antimicrobial susceptibility test results were interpreted according to the guidelines set by the Clinical and Laboratory Standards Institute (CLSI)26. Eight antibiotics, which were acquired from Thermo Fischer ScientificTM (Thermo Fisher, South Africa), were employed in this investigation: Ampicillin (10 µg), Ceftazidime (µg), Erythromycin (15 µg), Gentamicin (10 µg), Tetracycline (30 µg), Vancomycin (µg), Chloramphénicol (µg), and Streptomycine (10 µg). Antimicrobial resistance was determined using Enterococcus faecalis ATCC 29212 and E. faecium ATCC 700221 as the positive control. According to CLSI criteria, plates were interpreted as positive when growth of Enterococcus was observed within the inhibition zone adjacent to the antibiotic disk. Molecular detection of antibiotic-resistant genes The presence of genes associated with tetracycline, erythromycin, quinupristin/dalfopristin, and vancomycin was examined for each strain using specific PCR assays. Ten antibiotic-resistant genes were screened in this study, including tetracycline-resistant genes encoding ribosomal protection proteins (tetM and tetS) and efflux proteins (tetK and tetL), the transposon integrase gene (int) of the Tn916/Tn1545 family, erythromycin resistance gene (ermB), and vancomycin resistance genes (vanA and vanB) (Table 1). Evaluation of MAR index, MDR, and XDR patterns in bacterial isolates The formula described by Rajendiran et al.27, was used to calculate the multiple antibiotic resistance (MAR) index. MAR = (total number of antibiotics tested) / (number of antibiotics displaying resistance by an isolate). Values near 0 indicated more sensitivity, while values near 1 indicated significant resistance. A MAR score of 0.20 or higher suggested a significant resistance level or a high-risk reservoir of bacterial contamination. A high-risk reservoir of bacterial contamination in terms of MAR refers to an environment, host, or material that harbours bacteria and supports the persistence and dissemination of multidrug-resistant strains. Such reservoirs act as amplifiers and transmission hubs of resistant organisms, posing significant threats to public health and food safety. Additionally, MDR is characterized by nonsusceptibility to at least one agent in three or more categories of antimicrobials. Statistical analysis Descriptive statistics, particularly proportions, percentages, means, and two-way ANOVA and Fisher’s exact test, were calculated using Excel software (2013) with P ≤ 0.05 considered significant. Results California mastitis test (CMT) and somatic cell counts (SCC) Out of the 174 cows that were sampled, only 68 (39.0%) were positive for subclinical mastitis at a cow level based on CMT. Moreover, only 194/696 (27.8%) quarter-level samples were positive for subclinical mastitis based on the SCC assay. Identification and detection of Enterococcus spp. Isolates A total of 62 presumptive Enterococcus strains were isolated from 68 milk samples that were considered to have subclinical mastitis from 174 dairy cows across six farms. MALDI-TOF-MS analysis and dd1 species-specific gene PCR assays confirmed the identification of all 58 (93%) E. faecalis and 4 (6.4%) E. faecium isolates from milk samples. The 0.6% remaining were all negatives and did not amplify by species-specific genes. In contrast, the study isolated 34 presumptive Enterococcus species from cow dung, with all (100%) identified as E. faecalis. All confirmed isolates in this study are shown in Fig. 2. Furthermore, using Fisher’s exact test, our study found that there’s no statistically significant association between sample type (Milk vs. Cow Dung) and the presence of E. faecalis vs. E. faecium (p-value: 0.294). Fig. 2The number of Enterococcus species isolates cultured from milk and cow dung samples from dairy cows with subclinical mastitis. The number of Enterococcus species isolates cultured from milk and cow dung samples from dairy cows with subclinical mastitis. Phenotypic antimicrobial resistance The results of the antimicrobial resistance analysis using the disc diffusion method for E. faecalis and E. faecium isolates from bovine milk and cow dung are presented in Table 2. Among all 58 E. faecalis isolates identified from milk samples, 10 (17.2%) showed resistance to vancomycin, followed by tetracycline, ceftazidime, erythromycin, and streptomycin at 7 (12.0%), 6 (10.3%), 6 (10.3%), and 4 (6.8%), respectively. All E. faecium (100%) isolates from milk samples were resistant to vancomycin, with resistance also observed to streptomycin, erythromycin, and ceftazidime at 3 (75%), 2 (50%), and 2 (50%), respectively. Furthermore, the results indicated that 11 (32.3%) of the E. faecalis isolates from cow dung were resistant to vancomycin, followed by resistance to ceftazidime, tetracycline, and ampicillin, with 7 (20.5%), 5 (14.7%), and 5 (14.7%), respectively. The statistical analysis revealed that for both E. faecalis and E. faecium from milk samples p-value was 0.001, while isolates from cow dung samples had a p-value of 0.01, indicating statistically significant results between the number of isolates from each sample and the resistance proportion of antibiotics used. Table 2Antimicrobial resistance profiles of Enterococcus faecalis and E. faecium isolated from milk and cow dung samples.AntibioticsSamplesMilkCow dungE. faecalisE. faeciumE. faecalisE. faeciumVancomycin10 (17.2%)4 (100%)11 (32.3%)-Gentamycin2 (3.4%)-2 (5.8%)-Tetracycline7 (12.0%)-5 (14.7%)-Ceftazidime6 (10.3%)2 (50%)7 (20.5%)-Streptomycin4 (6.8%)3 (75%)1 (2.9%)-Chloramphenicol3 (5.1%)-4 (11.7%)-Ampicillin2 (3.4%)-5 (14.7%)-Erythromycin6 (9.6%)2 (50%)0 (0%)-p-value0.0010.01 Table 2Antimicrobial resistance profiles of Enterococcus faecalis and E. faecium isolated from milk and cow dung samples.AntibioticsSamplesMilkCow dungE. faecalisE. faeciumE. faecalisE. faeciumVancomycin10 (17.2%)4 (100%)11 (32.3%)-Gentamycin2 (3.4%)-2 (5.8%)-Tetracycline7 (12.0%)-5 (14.7%)-Ceftazidime6 (10.3%)2 (50%)7 (20.5%)-Streptomycin4 (6.8%)3 (75%)1 (2.9%)-Chloramphenicol3 (5.1%)-4 (11.7%)-Ampicillin2 (3.4%)-5 (14.7%)-Erythromycin6 (9.6%)2 (50%)0 (0%)-p-value0.0010.01 Antimicrobial resistance profiles of Enterococcus faecalis and E. faecium isolated from milk and cow dung samples. The multidrug-resistant profile of the E. faecalis and E. faecium isolates The results showed that 16 (25.6%) of E. faecalis and E. faecium isolates were classified as multidrug-resistant. Six (10.3%) E. faecalis and 6 (10.3%) E. faecium isolates from milk were categorized as MDR, respectively. Only 4 (6.8%) E. faecalis isolates from cow dung samples were MDR. The multidrug-resistance patterns of the isolates are shown below (Table 3). Table 3Multidrug-resistant profiles of E. faecalis and E. faecium isolates from milk and cow Dung samples.AMR phenotypesMilkCow dungE. faecalisE. faeciumE. faecalisCAZ, S, AM1--C, AM, E---TE, S, C-11VA, GEN, TE--1VA, TE, C-11VA, TE, CAZ1--VA, CAZ, S---VA, CAZ, AM114VA. TE, S, E---VA, TE, CAZ, S, E---VA, TE, C, AM1-1VA, GEN, TE, C121VA, GEN, TE, CAZ, E-1-VA, GEN, C, AM, E1--Total664 Table 3Multidrug-resistant profiles of E. faecalis and E. faecium isolates from milk and cow Dung samples.AMR phenotypesMilkCow dungE. faecalisE. faeciumE. faecalisCAZ, S, AM1--C, AM, E---TE, S, C-11VA, GEN, TE--1VA, TE, C-11VA, TE, CAZ1--VA, CAZ, S---VA, CAZ, AM114VA. TE, S, E---VA, TE, CAZ, S, E---VA, TE, C, AM1-1VA, GEN, TE, C121VA, GEN, TE, CAZ, E-1-VA, GEN, C, AM, E1--Total664 Multidrug-resistant profiles of E. faecalis and E. faecium isolates from milk and cow Dung samples. The MAR index of the resistant isolates Table 4 of the MAR results shows that the 19 Enterococcus isolates cultured from milk and cow dung exhibit significant antibiotic resistance. Table 4MAR index of the Enterococcus isolates from milk and cow dung.Isolate noNo. of antibiotics to whichisolate was resistant (a)MAR index = a/bM140.5M250.6M330.3M450.6M530.3M630.3M730.3M830.3M930.3M1050.6CD130.3CD240.5CD330.3CD430.3CD540.5CD630.3CD730.3CD830.3CD930.3 Table 4MAR index of the Enterococcus isolates from milk and cow dung.Isolate noNo. of antibiotics to whichisolate was resistant (a)MAR index = a/bM140.5M250.6M330.3M450.6M530.3M630.3M730.3M830.3M930.3M1050.6CD130.3CD240.5CD330.3CD430.3CD540.5CD630.3CD730.3CD830.3CD930.3 MAR index of the Enterococcus isolates from milk and cow dung. Molecular detection of antimicrobial-resistant genes The genomic detection of antimicrobial resistant genes (ARGs) revealed that the majority of all isolates harboured the vanA gene, with 60 (96%), followed by tetM, ermB, vanB, and tetL, each present in 20 (62.5%) isolates. Additionally, 13 (20.9%) isolates from milk samples carried the tetk gene. Of these isolates, 56 (93.3%) harbouring the vanA gene were E. faecalis, while 4 (6.6%) were E. faecium. Furthermore, 20 (62.5%) of the tetM isolates and 17 (85%) of the tetk isolates were E. faecalis, with the remaining 3 (15%) being E. faecium. The vanB gene was found exclusively in E. faecalis. The genomic results also showed that, from cow dung samples, vanA was the most frequently detected gene, present in 17 (50%) isolates. This was followed by vanB and tetM, with 13 (38.2%) and 5 (8.3%) isolates, respectively. None of the isolates from cow dung carried the ermB, tetL, tetS, vanD, or vanE genes. Interestingly, the transposon integrase gene (int) was detected in 29 (87%) E. faecalis strains and 4 (12.1%) E. faecium strains from milk samples. In comparison, 15 (46.4%) of the E. faecalis isolates from cow dung also showed the presence of the transposon integrase gene (int) (Fig. 3A and B). Fig. 3Heatmap illustrating the clustering of antimicrobial-resistant genes and virulence genes detected from cow dung (A) and milk samples (B) derived Enterococcus spp. isolates. Heatmap illustrating the clustering of antimicrobial-resistant genes and virulence genes detected from cow dung (A) and milk samples (B) derived Enterococcus spp. isolates. Molecular detection of virulence genes The results of the virulence profile further showed that 20 (33.3%) of the isolates from milk samples harboured the asa1 gene, followed by the ace and esp genes with 11 (12.7%) and 6 (10%) respectively. Moreover, none of the isolates exhibited detection of the hyl and cylA genes. Of the 20 isolates with the asa1 gene detected, 17 (85%) were E. faecalis, while 4 (20%) were E. faecium. Furthermore, 10 (90.9%) isolates with the ace gene were E. faecalis, and 1 (10%) was E. faecium. The esp gene was detected in both E. faecalis and E. faecium, with 3 (50%) each. In addition to milk samples, only two virulence genes, gelE and hyl, were detected in isolates from cow dung samples, with 33 (53.2%) and 13 (38.2%) isolates, respectively. A total of 30 (90.9%) isolates carrying the gelE gene were E. faecalis, while the remaining 3 (9.0%) were E. faecium. Moreover, all 13 (38.2%) isolates carrying the hyl gene were E. faecalis (Fig. 3A and B). Coexistence of phenotypic and genotypic antibiotic resistance among E. faecium and E. faecalis This study further showed that 8 out of 59 (13.5%) strains of E. Faecalis and 4 (100%) strains of E. Faecium from milk samples exhibited a coexistence of phenotypic and genotypic antimicrobial resistance (AMR) and virulence factors (Table 5). In contrast, all four strains (11.7%) of E. faecalis isolated from cow dung demonstrated the same coexistence of antimicrobial resistance and virulence factors (Table 6). Table 5Co-occurrences of resistance phenotypes, genotype, and the prevalence of Enterococcus strains from milk samples.E. faecalisE. faeciumIsolates ID:PhenotypicAMR genesVirulence genesIsolates ID:PhenotypicAMR genesVirulence genesMS2VA, TE, E, S, CAZ,vanA, tetM, ermB, vanB, tetL, tetK, intAsa1, ace, espMS35VA, S, E, CAZInt vanA, tetMEsp, Asa1MS3VA, TE, E, S, CAZvanA, tetM, ermB,vanB, tetL, tetK, intAsa1, aceMS36VA, S, E, CAZInt, vanA, tetM Asa1 MS4VA, TE, E, S, CAZAsa1, ace, espMS41VA, SInt, vanA, tetm Asa1 MS10VA, TE, E, CAZvanA, tetM, ermB, vanB, tetL, tetK, intAsa1, ace, espMS44CAInt, vanA Asa1 MS25VA, TE, E, CAZ,vanA, tetM, ermB, vanB, tetL, tetK, intAsa1, ace, esp-- - - MS26VA, TEvanA, tetM, ermB, vanB, tetL, tetK, intAsa1, ace,-- - - MS27VAvanA, tetM, ermB, intvanB,tetL, tetK, intAsa1, ace-- - - MS29VAvanA, tetM, ermB, vanB, tetL, tetK, intAsa1, ace-- - -*VA, TE, E, S, CAZ, GEN, C, AMP: Vancomycin, Tetracycline, Erythromycin, Streptomycin, Ceftazidime. Table 5Co-occurrences of resistance phenotypes, genotype, and the prevalence of Enterococcus strains from milk samples.E. faecalisE. faeciumIsolates ID:PhenotypicAMR genesVirulence genesIsolates ID:PhenotypicAMR genesVirulence genesMS2VA, TE, E, S, CAZ,vanA, tetM, ermB, vanB, tetL, tetK, intAsa1, ace, espMS35VA, S, E, CAZInt vanA, tetMEsp, Asa1MS3VA, TE, E, S, CAZvanA, tetM, ermB,vanB, tetL, tetK, intAsa1, aceMS36VA, S, E, CAZInt, vanA, tetM Asa1 MS4VA, TE, E, S, CAZAsa1, ace, espMS41VA, SInt, vanA, tetm Asa1 MS10VA, TE, E, CAZvanA, tetM, ermB, vanB, tetL, tetK, intAsa1, ace, espMS44CAInt, vanA Asa1 MS25VA, TE, E, CAZ,vanA, tetM, ermB, vanB, tetL, tetK, intAsa1, ace, esp-- - - MS26VA, TEvanA, tetM, ermB, vanB, tetL, tetK, intAsa1, ace,-- - - MS27VAvanA, tetM, ermB, intvanB,tetL, tetK, intAsa1, ace-- - - MS29VAvanA, tetM, ermB, vanB, tetL, tetK, intAsa1, ace-- - -*VA, TE, E, S, CAZ, GEN, C, AMP: Vancomycin, Tetracycline, Erythromycin, Streptomycin, Ceftazidime. Co-occurrences of resistance phenotypes, genotype, and the prevalence of Enterococcus strains from milk samples. vanA, tetM, ermB, vanB, tetL, tetK, int vanA, tetM, ermB,vanB, tetL, tetK, int vanA, tetM, ermB, vanB, tetL, tetK, int vanA, tetM, ermB, vanB, tetL, tetK, int vanA, tetM, ermB, vanB, tetL, tetK, int vanA, tetM, ermB, intvanB, tetL, tetK, int vanA, tetM, ermB, vanB, tetL, tetK, int *VA, TE, E, S, CAZ, GEN, C, AMP: Vancomycin, Tetracycline, Erythromycin, Streptomycin, Ceftazidime. Table 6Co-occurrences of resistance phenotypes, genotype, and the prevalence of Enterococcus strains from cow Dung samples.E. faecalisIsolates ID:PhenotypicAMR genesVirulence genesCD10VA, CAZ, TEvanA, VanB, tetM, intgelE, hylCD12VA, CAZ, TE, AMPVanA, vanB, tetM, intgelE, hylCD21VA, CAZ, TE, AMPvanA, vanB, tetM, intgelE, hylCD34VA, CAZ, TE, AMPvanA, vanB, tetM, intgelE, hyl, esp, ace*VA, TE, E, S, CAZ, GEN, C, AMP: Vancomycin, Tetracycline, Erythromycin, Streptomycin, Ceftazidime. Table 6Co-occurrences of resistance phenotypes, genotype, and the prevalence of Enterococcus strains from cow Dung samples.E. faecalisIsolates ID:PhenotypicAMR genesVirulence genesCD10VA, CAZ, TEvanA, VanB, tetM, intgelE, hylCD12VA, CAZ, TE, AMPVanA, vanB, tetM, intgelE, hylCD21VA, CAZ, TE, AMPvanA, vanB, tetM, intgelE, hylCD34VA, CAZ, TE, AMPvanA, vanB, tetM, intgelE, hyl, esp, ace*VA, TE, E, S, CAZ, GEN, C, AMP: Vancomycin, Tetracycline, Erythromycin, Streptomycin, Ceftazidime. Co-occurrences of resistance phenotypes, genotype, and the prevalence of Enterococcus strains from cow Dung samples. *VA, TE, E, S, CAZ, GEN, C, AMP: Vancomycin, Tetracycline, Erythromycin, Streptomycin, Ceftazidime. Discussions A comparison with earlier studies conducted in similar management systems in Sri Lanka, which reported SCM prevalence rates of 43.0% at the cow level and 49.0% at the quarter level28,29, reveals that the current study found slightly lower prevalence rates of 39.0% at the cow level and 27.8% at the quarter level, respectively. Nonetheless, similar prevalence levels were discovered in crossbred dairy cow herds in Bangladesh that were managed using a semi-intensive system30 and in equally hot tropical parts of Mexico31. In contrast to the current study, other research has found higher prevalence rates of SCM in smallholder dairy farms with both grazing and zero grazing in Kampala, Uganda32, as well as under intensive management systems in southern Brazil33 and Poland34. Because of risk factors such low grazing, poor udder hygiene, increased parity, and late lactation, Abrahmsén et al.32, found an abnormally high prevalence of SCM in Uganda (86.2%). Although the incidence of SCM in Zimbabwe’s smallholder dairy farms was relatively low (16.3%)35, defined SCM as having SCC of at least 300 × 103 cells/mL. Enterococci are one of the environmental pathogens responsible for bovine mastitis, and antimicrobials are widely used to treat these infections9. Unfortunately, Enterococcus has inherent resistance to many broad-spectrum antibiotics, which poses a significant threat to public health. Understanding the enterococci’s prevalence, antimicrobial resistance, and virulence factors is crucial for targeted therapy and control of bovine mastitis. These findings align with research showing that E. faecalis is more frequently isolated from animal sources and dairy environments than E. faecium36,37. According to Montealegre et al.38, the prevalence of E. faecalis in cow manure may be due to its adaptation to the cattle’s digestive tract, where it persists as a commensal microbiota. Since E. faecalis and E. faecium are opportunistic pathogens capable of surviving in various conditions, their presence in milk samples may indicate contamination during milking or subsequent handling7. Enterococcus isolates from milk and cow dung samples exhibited distinctive patterns in their virulence gene profiles, which could indicate their ability to cause disease and adapt to their environment. The asa1 gene was most common among milk isolates, present in 33.3% of samples, mainly E. faecalis (85%). According to Gilmore et al.39, the asa1 gene encodes an aggregating substance that boosts colonisation and infection capabilities by encouraging adhesion and biofilm development. This prevalence emphasizes the significance of the asa1 gene in the pathogenesis of infections in dairy cattle. Saeed et al.40, noted that similar levels of asa1 were found in E. faecalis from the European gut microbiome of healthy individuals. The ace gene, which encodes a collagen-binding adhesin necessary for tissue attachment and persistence, was found in 12.7% of milk isolates (mostly E. faecalis) in Germany41. The esp gene, distributed equally between E. faecalis and E. faecium and present in 10% of milk isolates, is well known for its role in immune evasion and biofilm formation in Serbia42. While a study by Rahman et al.43, in Bangladesh has reported the presence of cylA and hyl genes in bovine-derived enterococci, their absence in milk isolates may suggest that the distribution of virulence genes varies depending on regional factors or management practices. Only the virulence genes gelE (53.2%) and hyl (38.2%) were found in cow dung samples. The hyl encodes hyaluronidase, facilitating bacterial spread through host tissues, while gelE encodes gelatinase, aiding tissue breakdown and immune evasion44. The nearly exclusive presence of both genes in E. faecalis suggests that this species is more virulent and better adapted to its environment than E. faecium45. Manure may serve as an environmental reservoir for virulent enterococci, potentially posing risks to both animal and public health, as evidenced by detecting these genes in dung7. The findings show that Enterococcus isolates from dairy sources, especially E. faecalis and E. faecium, have an antimicrobial resistance pattern. Notably, isolates from milk and cow dung had isolates which were resistant to vancomycin in this study. This is consistent with recent research showing that E. faecium is more often multidrug-resistant than E. faecalis46. Serious worries regarding zoonotic transmission through the food chain and environmental exposure are raised by the high level of resistance to vancomycin, a last-resort antibiotic in human medicine7. The extensive use of tetracycline, erythromycin, and streptomycin in cattle management, particularly in developing nations with less stringent regulations, is also reflected in the observed resistance to these antibiotics47. Due to their historical usage of tetracycline and macrolides (such as erythromycin) medicinal drugs, Enterococcus species isolated from dairy cattle frequently resist them48. Antimicrobial-resistant Enterococcus in cow dung points to an environmental reservoir of resistance genes that may be made worse by inappropriate manure disposal and ecosystem damage37. The discovery of ampicillin and ceftazidime resistance emphasizes the development of resistance mechanisms, which may have occurred due to horizontal gene transfer between bacterial genera or enterococci49. As previously noted, the vanA gene is the primary genetic factor responsible for vancomycin resistance in enterococci isolated from animal sources. The high prevalence of vanA, particularly among E. faecalis (93.3%) compared to E. faecium (6.6%), supports these findings39. Interestingly, the vanB gene was only found in isolates of E. faecalis. This aligns with research by Arredondo-Alonso et al.50, which showed that vanB carriage is less consistent and often linked with Enterococcus strains from Dutch hospitals. The widespread use of tetracyclines in veterinary medicine, which exerts selection pressure favouring the retention of such resistance determinants, is consistent with the high detection rates of tetracycline resistance genes tetM (62.5%) and tetK (20.9%) in milk isolates as shown by a global study of Barel et al.48. Nowakiewicz et al.51 in Poland observed that the co-occurrence of the tetM gene and the int integrase gene from the Tn916/Tn1545 transposon family in many E. faecalis strains highlights the role of mobile genetic elements in promoting horizontal gene transfer of tetracycline resistance. This was present in 87% of milk samples and 46.4% of dung samples. It is widely recognized that this integrative conjugative element facilitates the spread of resistance within and between bacterial populations in cattle contexts49. Interestingly, ermB, which causes resistance to macrolides, was found in milk isolates at a frequency comparable to tetM but not in dung isolates. Agga et al.52, observed niche-specific ARG distributions in cow farms, which could reflect variations in antibiotic exposure or selection pressures in different habitats. The absence of several ARGs (ermB, tetL, tetS, vanD, and vanE) in cow dung isolates highlights differences in the resistome between host-associated and environmental microbiota, possibly due to manure management techniques37. These findings support the idea that resistance genes form a dynamic and complex reservoir in dairy farm settings, with implications for spreading AMR to the broader environment and public health43. The MDR to antibiotics such as vancomycin, tetracycline, ceftazidime, erythromycin, and streptomycin was found in 20.6% of Enterococcus isolates from milk samples and 6.8% from cow dung. These findings are consistent with recent developments in veterinary microbiology. Studies conducted worldwide have raised concerns regarding treatment effectiveness and zoonotic transmission, showing the rising incidence of MDR enterococci in animals used for food production. For example, the 20.6% MDR rates discovered in milk isolates in this study closely matched the 25% MDR rates found in Enterococcus isolates from bovine mastitis in a survey by Saeed et al.40. Similar to but marginally higher than the 6.8% MDR found in cow dung samples in the current investigation, Rahman et al.43, reported MDR rates ranging between 10 and 15% in enterococci from cattle manure in Bangladesh, which may be due to regional or farm management variations. The MAR indices are routinely calculated to assess health concerns, particularly when antibiotic use is rare, and the risk of transmission to the community could have significant health implications53. In this study, the MAR indices were much higher than expected, indicating that treating infections associated with these strains is highly challenging. Similar to the findings of Adeyemi et al.54, most organisms exhibited values above 0.2, suggesting that these isolates are indeed high-risk pathogens originating from environments with frequent antibiotic use. Additionally, this investigation demonstrated that virulence factors and phenotypic and genotypic antimicrobial resistance (AMR) coexisted in 8 out of 59 (13.5%) strains of E. Faecalis and 4 (100%) strains of E. Faecium from milk samples. On the other hand, all four of the E. faecalis strains (11.7%) that were isolated from cow dung showed the similar coexistence of virulence factors and antibiotic resistance. Furthermore, given that some of these antibiotic-resistant medications work well against isolates that form biofilms, it is intriguing that asa1, gelE, and esp genes have a positive correlation with resistance profiles55. This is why we believe that exposure to antibiotic-resistant medications may cause biofilm-forming isolates to develop a particular resistance to these substances. Antibiotic-resistant bacteria benefit from the connection of asa1, esp, and gelE since it gives them two weapons to cause and survive clinical infection treatment55. Additionally, the findings demonstrated the discrepancy between genotypic and phenotypic AMR. Notwithstanding the benefits of molecular approaches, the capacity of molecular panels to forecast susceptibility is entirely reliant on their target list and our comprehension of resistance processes. Because the mechanism of AMR may not be included as a target on molecular panels, the absence of an AMR marker does not always indicate susceptibility for certain antimicrobials56. The majority of clinically meaningful resistance in Gram-positive organisms is often caused by a single mechanism of resistance (vanA/B for VRE), and forecasts of susceptibility profiles have been reported to have accuracy rates ranging from 98 to 100%57,58. Risk factors for subclinical mastitis, including advanced age and higher parity, specific udder and teat conformations, such as enlarged udders, poor teat end health (hyperkeratosis), and high udder dirtiness, were not taken into account in the current study. High humidity, poor herd cleanliness and sanitation, poorly maintained or insufficient milking equipment, particular housing systems, lengthier lactation stages, and inconsistent or insufficient post-milking teat disinfection are some environmental and management factors that enhance risk. Conclusion This study enhances our understanding of the enterococcal communities in raw milk and cow dung. Our findings indicate that cow dung does not serve as a source or potential reservoir for antimicrobial-resistant (AR) enterococci in the milk microbiota, as these microorganisms originate from different habitats than those found in milk. We hypothesize that farm surfaces or milking equipment may be the primary sources of enterococci in milk. It is also essential to consider the significant presence of multidrug-resistant (MDR) isolates and their capacity to spread AR genes through dairy products. Our research confirms a notable prevalence of vancomycin-resistant enterococci (VRE) in the dairy environment of South Africa, which is evident from the substantial presence of VRE in both milk and dung samples. To the best of our knowledge, this study is the first to investigate the relationship between the origin, pathogenicity, and antibiotic resistance of enterococci isolated from the milk of cows with subclinical mastitis and the corresponding bovine dung in the Free State Province, South Africa.