Although enterococci are considered sensitive to vancomycin, a rise in strains with resistance genes of potential transmission has been recorded (Tavares, 2014; Raza et al., 2018). These strains are termed Vancomycin-Resistant Enterococci (VRE) and are considered ‘superbacteria’ as determined by genotypic and phenotypic features (Werner, 2012; Tavares, 2014).

The importance of determining these genotypic and phenotypic profiles lies in the understanding of the movement of resistance genes, a phenomenon produced by plasmids and transposons in which genes are transferred across enterococci and other Gram-positive bacteria (Aun et al., 2021; Van Schaik et al., 2010; Werner, 2012).

Moreover, enterococci present relevance as an etiological agent of bovine mastitis. In a recent study performed in Poland with 2,000 mastitic milk samples from confirmed cases of bovine mastitis, 21.3% were caused by Enterococcus species (Ró?a?ska et al., 2019). Such inflammatory disease represents a major issue in dairy farming, especially when caused by VREs, which are more difficult to treat and can represent a huge risk of dissemination of resistance genes to other pathogens that causes mastitis, like S. aureus (Werner, 2012). Besides, VRE-induced mastitis can be a risk to public health as a foodborne pathogen, since VREs may be present in subclinical cases of mastitis and, therefore, infect humans through consumption of contaminated milk from apparently healthy cows (Werner, 2012; Kateete et al., 2013).

In Brazil, the most studied and prevalent bacterium in cattle is S. aureus. A recent study analyzed 400 strains isolated from mastitic animals in four Brazilian states, and showed that these strains have important resistance genes in almost all classes of antimicrobials currently used for human and animal treatment in the country (Pérez et  al., 2020). In the case of Klebiseila pneumoniae, another very prevalent bacterium in the country, the presence of genes encoding β-lactamases was observed in several bacteria, including genes for drug resistance for commonly used drugs in the treatment of mastitis in Brazilian dairy herds (gentamicin, cephalosporins, sulfamethoxazole-trimethoprim, tetracycline), as well as antimicrobials of critical importance to human health (meropenem, ceftazidime, fluoroquinolones) (Nobrega et al., 2021). A recent study, observed the species diversity and antimicrobial resistance patterns of Enterococcus spp. isolated from mastitis cases, milking machine and dairy cow environment. As a result, of the 365 isolates studied from several Brazilian municipalities, 1.9, 0.3 and 0.6% were resistant to penicillin, vancomycin and teicoplanin, respectively. This is a recent data that demonstrates the current situation (Juliano et al., 2022).

The relevance in assessing the multiresistance of enterococci isolated from bovine milk is in four main premises: (1) human transmission through ingestion of contaminated milk, which has paramount importance in Brazil where 20 to 30% of milk and dairy products are sold without inspection or heat treatment (IBGE, 2015); (2) significant gaps are found in the literature with respect to the multiresistant profile of enterococci isolated from bovine milk in Brazil; (3) Enterococcus spp. are neglected etiological agents of cattle mastitis (Kateete et al., 2013); and (4) Enterococcus spp. can transmit resistance genes to S. aureus, the most important pathogen of mastitis in cattle (Werner, 2012).

Therefore, the aims of this study were to determine the in vitro antimicrobial susceptibility profile of Enterococcus spp. isolated from bulk tanks and mastitic milk samples from dairy farms in the states of São Paulo and Minas Gerais (Brazil).

The farm

A total of 310 species of Enterococcus spp. were studied, of which 52 were from confirmed cases of bovine clinical mastitis and 258 from bulk tanks. The milk samples positive for enterococci were obtained between 2018 and 2019 from four dairy farms in São Paulo State and six in Minas Gerais State, Brazil.

All ten cattle herds had mastitis control programs, including computer software for data record, a somatic cell count (SCC) in bulk tanks up to 400,000 cs mL^-1, a minimum of 200 lactating cows, and an automatic milking machine. In addition, the milk from the dairy farms was supplied by a high-production Holstein Friesian cattle (> 20 L each animal per day).

Animals with mastitis were treated with different classes of antibiotics, depending on the farm of origin. The most used antibiotics, in descending order, were: ceftiofur (third generation cephalosporin), tetracycline hydrochloride (tetracycline class), neomycin (aminoglycoside class), bacitracin (aminoglycoside class), cefquinoma (cephalosporin class), cefoperazone (cephamycins class) and enrofloxacin (quinolones class).

Isolation

The cases of clinical mastitis were based upon detection of macroscopic changes common to mastitic milk (e.g. clumps, pus or blood) and the systemic signs of mastitis in dairy cows (e.g. fever, loss of appetite or behavior changes) (Radostits et al., 2007). Mastitic milk samples from these animals were collected in sterile vials following teat disinfection with 70% alcohol and were immediately kept refrigerated (maximum of 7°C) during transfer to the Research Nucleus on Mastitis (NUPEMAS) laboratory of the São Paulo State University (UNESP), School of Veterinary Medicine and Animal Science, Botucatu, São Paulo State, Brazil. At the same time, monthly bulk tanks milk samples of each farm were transferred to the laboratory with the same sterility and refrigeration conditions.

In the laboratory, mastitic milk samples were spread on MacConkey agar and blood agar plates, with the latter being supplemented with 5% defibrinated ovine blood. Samples were incubated at 37°C under aerobic conditions, and examined every 24 h for three days. The bacteria were classified according to the criteria adopted by the National Mastitis Council (NMC, 1999). The colonies that were morphologically compatible with enterococci were analyzed after Gram stain by microscopy, as well as by catalase and hemolysins tests. The samples were assayed by the following biochemical tests as described by Facklam and Collins (1989) and updated by Facklam (2007): growth in halophyte broth (6.5% NaCl), esculin hydrolysis, hydrolysis of L-pyrrolidonyl-beta-naphthylamide (PYR), arginine decarboxylase test, pigment production, motility, tetrazolium reduction, mannitol, arabinose, raffinose and sorbitol fermentation test.

From the bulk tanks, 1 mL of sample was pipetted into test tubes along with 9 mL of 0.1% sterile peptone water (CLSI, 2014). Later, dilutions were carried out in a serial fashion from 10^-1 to 10^-3. A total volume of 1 mL from each dilution was spread onto Enterococcosel agar, and incubated at 37°C for 24 h to identify the enterococci colonies based on their morphological characteristics.

In vitro antimicrobial susceptibility profile

As proposed by Bauer et al. (1966), the antimicrobial susceptibility was evaluated by the disk diffusion method, where the bacteria were submitted to the following antimicrobials in Mueller-Hinton agar: Ampicillin - 10 µg (AMP), Cephalexin - 30 µg (CL), Cefoxitin - 10 µg (CFO), Ciprofloxacin - 10 µg (CIP), Enrofloxacin - 10 µg (ENR), Gentamicin -10 µg  (GEN), Marbofloxacin - 5 µg (MRB), Neomycin - 30 µg (N), Novobiocin - 5 µg (NOV),  Oxacillin - 10 µg (OXA), Penicillin G - 10 U.I. (PEN), Sulfamethoxazole/Trimethoprim - 25 µg (SXT), Tetracycline - 30 µg (TE), Teicoplanin - 30 µg (TEC) and Vancomycin - 10 µg (VAN). International reference strains were used with a positive control (E. faecium BM 4147 vanA genotype) and a negative control (S. aureus ATCC 25923). The zone of inhibition was interpreted according to CLSI (2014), and ascribing the isolates the categories susceptible (S), intermediate (I) and resistant (R).

MIC test of milk samples

The intermediate enterococci samples to VAN were tested with the minimum inhibitory concentration (MIC) method, which exhibits a higher accuracy than the disk diffusion method (Bauer et al., 1966). The commercial kit E-test^TM was used in Mueller-Hinton agar, following the manufacturer's recommendations (Bauer et al., 1966; CLSI, 2014).

Multiresistance classification

The enterococci isolates that were simultaneously resistant to at least three different groups/classes of antimicrobials were designated multi-resistant.

Statistics

Descriptive statistics were used through the distribution of absolute frequencies relative to the results obtained in the in vitro microbial sensitivity tests. Fisher´s exact test was performed to verify if there were differences between the percentage of multiresistant strains from bulk tanks and mastitis samples. Kruskal Wallis test was used to verify if the median number of antimicrobial compounds to which the isolates were resistant differed between bulk or mastitis isolates. The significance level of 0.05 was considered for the analyses.

Resistance to fluorquinolones was low in strains from bulk tanks and clinical mastitis (less than 10% of isolates). However, there was a high percentage of intermediate resistance in the isolates from bulk tanks, 54.7, 77.1, and 66.3%, respectively to CIP, ENR, and MRB. Of the clinical mastitis isolates, 11.5 and 32.7% were resistant to CIP and MRB, respectively (Table 1). And for CIP, Ró?a?ska et al. (2019) noticed 0.47% of enterococci resistant from clinical mastitis.  

Fabianová et al.  (2010) reported 8.7 and 4.7% of resistant enterococci from bulk tanks to GEN and AMP, respectively. In this study, a low percentage of resistant strains were observed to AMP, only 0.8%. For GEN, 29.8% were resistant, while 29% presented an intermediate resistance (Table 1).  

Erbas et al. (2016) noticed a high prevalence of resistant enterococci to TE, 81% of isolates isolated from clinical mastitis, while Ró?a?ska et al. (2019) observed 61.5%.  In this study 57.7% of resistant strains were identified as shown in Table 1. 

According to Fisher's test, the percentage of multidrug-resistant strains in the tanks was 94.96% and in the animals was 90.38%. There was no statistical difference, with a p value of 0.1986. According to the Kruskal Wallis test, the median number of resistant compounds of the tank was equal to 6, and the median number of resistant compounds of the animals with mastitis was 5. There was no statistical difference in the median of the two groups, and the p value was 0.1350. In summary, there was no considerable statistical difference between isolates originating from bulk tanks and from animals with mastitis. We suggest that milk samples collected from the tanks may be adequate for antimicrobial sensitivity profiling research, as they are simpler and less expensive for farmers.

Data about antimicrobial sensitivity of Enterococcus spp. from milk from cattle, in bulk tanks or clinical mastitis, are scarce. Studies related to this topic indicate that there are many factors involved in the emergence of resistant strains, and that this will depend on the farm, the management performed, and the antimicrobial treatment applied (Ró?a?ska et al., 2019; Erbas et al., 2016; Fabianová et al., 2010).

In Brazil, about 20 to 30% of milk and dairy products are sold without any heat treatment (IBGE, 2015). Our results show that more than 90% of enterococci from bulk tanks, which are milk ready for use are multiresistant, indicating risks to public health through ingestion of contaminated milk. In addition, the presence of these enterococci infecting cows could lead to horizontal transfer of resistance genes to other Gram-positive bacteria present in the mammary glands.

The treatment and the preventive use of antimicrobials in bovine mastitis cases are associated with the indiscriminate use of these drugs, and probably this is the main reason for the increase of resistant strains in cattle

farms (Krömker and Leimbach, 2017). Many studies connect the indiscriminate use of antibiotics in animal production with the dissemination of multiresistant Enterococcus spp. Therefore, the transmission of these pathogens by foods is a big risk to public health (Tavaras et al., 2012; Ali et al., 2013; Ró?a?ska et al., 2019).

More studies are needed to identify the risk factors associated to multiresistant Enterococci spp. causing bovine mastitis, including the use of genetic analyses able to characterize the enterococci species. This study might assist future researches on the impact of enterococci in animal and public health, as well as guide which treatment protocols can be used in cases of bovine mastitis.  

No Enterococcus spp. resistant to vancomycin was detected. However, a high multiresistant enterococci strains were detected in bulk tanks (95%) and from clinical mastitis cases (90.4%), pointing to the risks of misuse of antibiotics in dairy cattle farming. These data highlight the need for more studies on the impacts of enterococci infections in both human and animal health, especially towards the possibility of transmission of resistance genes from Enterococcus spp. to other pathogenic bacteria.

The authors have not declared any conflict of interests.

This study was funded by the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), processes number 2015/19688-8, 2017/08823-7, 2018/09103-0.