Introduction

Bovine mastitis, characterized by an inflammatory reaction in the mammary glands, is considered one of the major diseases that affect dairy cows. It can cause several productive and economic losses, such as reduced production of milk, increased expenses for the treatment, and in some cases, early disposal of animals [1,2,3,4].

Regarding the causative agents of bovine mastitis, the genus Staphylococcus is one of the most researched pathogens regarding the etiology of bovine mastitis, especially to subclinical mastitis. The importance of this genus is associated with its high resistance to antimicrobials, enterotoxaemia cases in humans, and disease development in animals [5,6,7]. Initially, studies on bovine mastitis had focused on S. aureus. However, recently, there has been an increase in research related to resistant coagulase-negative staphylococci (CoNS), with emphasis on Staphylococcus epidermidis and Staphylococcus chromogenes [8,9,10,11,12].

In recent years, antimicrobial resistance has become a major problem for the treatment of both animal and human diseases, as it is already possible to identify several species of bacteria carrying genes that confer resistance to some antimicrobials, such as the blaZ, mecA, and mecC (resistance to β-lactams) [13,14,15], tetL and tetM (tetracycline resistance) [13, 16], and vanA and vanB (vancomycin resistance) [17]. For example, gram-positive bacteria, such as Staphylococcus spp., can transmit these genes horizontally to other Staphylococcus spp. and/or other gram-positive bacteria, whether of human or animal origin [16].

Several studies have investigated the etiology of bovine mastitis and the resistance profile of the causative agents in various Brazilian states, such as Rio de Janeiro [18], Pará [19], São Paulo [20], Minas Gerais [13], Pernambuco [21, 22], Mato Grosso [23], and Paraná [10]. However, there are still various lacunes in other states, such as Piauí state, where there are no recordings of the said illness in the dairy region, nor records of the resistance profile of bacteria causing bovine mastitis. Likewise, this research aimed to identify the diversity of bacterial species of the genus Staphylococcus spp. in subclinical mastitis in dairy herds in the state of Piauí, Northeastern Brazil, and to evaluate the phenotypic and genotypic resistance profile.

Material and methods

Sampling and sample collection

Farms were chosen non-probabilistically for convenience. Milk samples were collected from 17 farms, being 3 in the municipality of Luiz Correia, 3 in Buriti dos Lopes and 11 in the municipality of Parnaíba, both municipalities located in the dairy basin of Piauí, Northeast Brazil. Initially, the California Mastitis Test (CMT) was performed on 680 breast quarters (170 Girolando cows) as per the protocol provided by Schalm and Noorlander [24]. The teats had been previously sanitized for subsequent testing. Milk samples were collected from the glands that displayed two (+ +) or three (+ + +) crosses in the CMT result, totaling 321 positive samples. Subsequently, these samples were stored in sterile bottles, placed in isothermal boxes, refrigerated, and sent for microbiological and molecular analyses in the Laboratory of Animals Infectious Diseases of the Federal Rural University of Pernambuco.

Microbiological isolation and identification of Staphylococcus spp.

Milk samples were cultivated in Base Agar supplemented with 7% sheep blood. The plates were incubated in a bacteriological incubator at 37 °C for 72 h and evaluated every 24 h. In post bacterial growth, the colonies were characterized by their morphology and morphotintorial characteristics via Gram staining, followed by biochemical tests of catalases in Staphylococcus spp. Subsequently, species identification was performed by the matrix-assisted laser desorption ionization time-of-flight mass spectroscopy (MALDI-TOF MS) (Bruker Daltonics) of the Institute of Pharmacology and Molecular Biology of the Federal University of São Paulo, as described by Wolters et al. [25].

Evaluation of antimicrobial resistance in Staphylococcus spp.

The phenotypic profile of antimicrobial resistance in Staphylococcus spp. was determined using the agar disk diffusion method [26]. The antimicrobials used were ampicillin (10 µg), penicillin G (10 UI), cefoxitin (30 µg), gentamicin (10 µg), oxacillin (1 µg), tetracycline (30 µg), erythromycin (15 µg), and vancomycin (30 µg), as per the recommendations in the Clinical and Laboratory Standards Institute guidelines [27].

Evaluation of resistance genes in Staphylococcus spp.

To obtain bacterial DNA, Staphylococcus spp. colonies were subjected to the DNA extraction method described by Fan et al. [28]. The genotypical resistance profile was evaluated for the genes tetM and tetL for tetracycline, blaZ for penicillin, vanA and vanB for vancomycin, and mecC and mecA for methicillin (Table 1).

Table 1 Genes, oligonucleotide sequences, size of amplified fragments, and reference
Full size table

Conventional polymerase chain reaction (PCR) was employed to amplify specific regions of these genes according to their thermal profiles with some modifications in the reagent concentrations. To this end, the final volume of each reaction was 12.5 µl, containing 100 ng of template DNA, 10 pmol of forward and reverse primers, and 6.25 µl of Go-TaqGreen Master Mix (Promega). Bacterial strains harboring these genes were used as a positive control, and ultrapure Milli-Q water was used as a negative control. The PCR products were stained with Blue Green (LGC Biotechnology) and subjected to electrophoresis in 1.5% agarose gel for 50 min at 100 V. The separated DNA bands were visualized and photographed by a photo documenter under ultraviolet light.

Statistical analysis

The results of microbiological analysis, polymerase chain reaction, and disk diffusion technique were expressed in relative and absolute frequencies [35].

Results

In this study, 57.63% of the CMT-positive samples (185/321) were also positive in the microbiological examination, and 59.45% (110/185) of the positive isolates obtained belonged to Staphylococcus spp. The MALDI-TOF MS analysis detected the following Staphylococcus spp. among the positive isolates: S. aureus, 68.1% (75/110); S. chromogenes, 12.7% (14/110); S. epidermidis, 5.4% (6/110); Staphylococcus sciuri, 4.5% (5/110); Staphylococcus warneri, 2.7% (3/110); Staphylococcus haemolyticus, 1.8% (2/110); Staphylococcus hominis, 1.8% (2/110); Staphylococcus arlettae, 0.9% (1/110); Staphylococcus capitis, 0.9% (1/110); and Staphylococcus gallinarum, 0.9% (1/110).

The results of the phenotypic resistance test, evaluating the resistance of Staphylococcus spp. to antimicrobials, demonstrated that 70.0% (77/110) and 61.8% (68/110) of Staphylococcus spp. isolates were resistant to penicillin and ampicillin, respectively. On the other hand, only 10.9% (12/110) and 11.8% (13/110) of the isolates were resistant to gentamicin and vancomycin, respectively. The distribution of phenotypic resistance of Staphylococcus species against each antimicrobial tested is described in Fig. 1.

Fig. 1
figure 1

Relative frequencies of phenotypic antimicrobial resistance for different species of Staphylococcus. Color variations show different percentages. 0 = no resistant samples, (AMP) ampicillin, (CFX) cefoxitin, (ERI) erythromycin, (GEN) gentamicin, (OXA) oxacillin, (PEN) penicillin, (TET) tetracycline, (VAN) vancomycin

Full size image

The genotypic resistance test in Staphylococcus spp. revealed the presence of the blaZ gene in 60.9% (67/110) of the isolates and the tetL and tetM in 20% (22/110) of the isolates, each. Additionally, 0.9% (1/110) of the isolates presented the mecA and vanB genes, while none of them had the mecC and vanA genes (Fig. 2).

Fig. 2
figure 2

Relative frequency of resistance genes for different species of Staphylococcus. Color variations show different percentages. 0 = absence of resistance gene

Full size image

Of the samples carrying the blaZ gene, 83.5% (56/67) showed phenotypic resistance to penicillin; 63.6% (14/22) of the tetL positive samples were resistant to tetracycline in the disc-diffusion test, and 59.0% (13/22) of the positive tetL samples were also phenotypically resistant to tetracycline. One positive mecA sample (0.9%; 1/110) also presented phenotypical resistance to oxacillin and to cefoxitin, and one vanB positive sample (0.9% 1/110) was resistant to vancomycin in the phenotypical test.

Among the Staphylococcus spp. isolates, 10.9% (12/110) had both tetracycline resistance genes (tetL and tetM) and 7.2% (8/110) presented multiple genotypical resistance, presenting three resistance genes (blaZ, tetL, and tetM).

Discussion

This is the first study evaluating the occurrence of bovine subclinical mastitis caused by Staphylococcus spp. and investigating the phenotypic and genotypic resistance profile of these bacteria in Piauí, Brazil. Our findings concurred with that of most etiological studies (those conducted in Brazil and globally) on bovine subclinical mastitis that Staphylococcus spp. were predominant (59.45%; 110/185) in the milk samples extracted from cows affected with subclinical mastitis [7, 8, 12, 19, 21, 36,37,38,39]. This study can be considered the most widespread study in identifying all Staphylococcus spp. (S. arlettae, S. capitis, S. chromogenes, S. gallinarum, S. haemolyticus, S. hominis, S. sciuri, and S. warneri) responsible for causing bovine subclinical mastitis in the Northeast region of Brazil; most studies have only identified S. aureus and coagulase-negative staphylococci (CoNS) as the causative agents in this region [21, 22, 36, 39]. These species, while never identified in Northeast Brazil, have been identified in the South and Southeast regions of Brazil [38]. Additionally, our findings emphasize the importance of the MALDI-TOF MS technique in investigating the etiology of bovine subclinical mastitis, as it is quick, cost-efficient, and easily operatable [40].

Post identifying the bacterial species, we generated the phenotypic resistance profile of the Staphylococcus spp. isolates using the disk diffusion test. We observed that resistance to penicillin and ampicillin was above 60.0%. This observation can be attributed to the wide-scale use of these drugs in the treatment of this disease [21, 41]. On the other hand, less than 40.0% of the isolates displayed resistance against other antimicrobials.

Several factors are pointed out as possible causes of the emergence of bacterial isolates resistant to antimicrobials in the agricultural production environment, among them the high use of antimicrobials or their indiscriminate use stand out [42, 43]. Both characterize the acquired form of resistance, in which a bacterial population that was naturally susceptible to the antimicrobial becomes resistant due to mutations in chromosomal genes or due to the acquisition of external genetic determinants of resistance [44]. Particularly for beta-lactams (such as penicillin and ampicillin), antimicrobials with the highest percentage of resistant isolates detected in this study, two mechanisms responsible for resistance are frequently reported. The first is the production of enzymes that inactivate antimicrobials, resulting in the destruction of the beta-lactam ring; and the second is the modification of the antimicrobial target, causing a decrease, or total loss, of the affinity between the drug and its binding site [15, 45, 46]. These mechanisms were studied and detected in the present study, demonstrating their presence in Staphylococcus spp. that causes of subclinical mastitis in the state of Piaui.

Regarding gentamicin, 10.9% of the isolates were resistant, this low resistance of the Staphylococcus species isolated in this study may be related to its little use in dairy cattle, due to its toxic potential for animals and the prolonged residual power in milk [47, 48] and according to Awandkar et al. [49], the low preference to gentamicin in veterinary therapy may be the reason behind this high sensitivity.

Regarding phenotypic resistance and the search for penicillin resistance genes, 70% (77/110) of Staphylococcus spp. were resistant in the disk-diffusion test, and of these, 87.0% (67/77) carried the blaZ gene, being 64.0% (48/75) of the S. aureus species and 45.7% (16/35) of the CoNS group. This high number of blaZ-carrying Staphylococcus spp. have already been observed in studies conducted in Brazil, such as Krewer et al. [50], with 93.1% (203/218), Martini et al. [13] with 97.7% (88/90), Santos et al. [21] with 68.9% (111/161), and Silva et al. [22] with 74.07% (20/27), and in other regions of the world such as the USA, with 53.48% (46/86) by Ruegg et al. [51] and in China, with 94.6% (35/37) by Yang et al. [52]. Only S. arlettae, S. capitis, and S. gallinarum did not presented the blaZ gene. The gene blaZ increases the production of β-lactamases in a cell; thus, inactivating β-lactams and conferring resistance against these compounds in bacteria [53]. Resistance to β-lactams can be mainly attributed to the indiscriminate use of these antimicrobials in mastitis treatment [21] and the increased occurrence of the blaZ gene in several Staphylococcus spp. as was observed in this study.

Concerning the genes that confer resistance against tetracycline, 20.0% (22/110) of the Staphylococcus spp. isolates were positive for the tetL gene and 20.0% (22/110) for the tetM gene. In S. aureus, the frequency was 13.3% (10/75) for tetL and 9.3% (7/75) for tetM. A higher frequency was observed in the CoNS group, with 35.2% (12/35) for tetL and 42.8% (15/35) for tetM, with emphasis on S. chromogenes and S. sciuri. This study is the first in reporting the occurrence of tetracycline resistance genes tetL and tetM in CoNS in bovine subclinical mastitis isolates in the Northeast region of Brazil. Presently, there are only a few studies reporting these in S. aureus; however, they are restricted to the Southeast region of Brazil [13, 16], where frequencies of occurrence of these genes were observed to be 8.8% for tetL and 2.2% for tetM [13] and 1.61% for tetL and 3.22% for tetM [16]. It is important to highlight that the tetL and tetM have distinct mechanisms, one, caused by the tetL gene, is an antimicrobial efflux system, and the other, caused, by the tetM gene, caused ribosome protection, this portrays the bacterial versatility regarding resistance acquisition [30].

We found that only one (1/110) isolate of S. epidermidis harbored the mecA gene. To date, only one related study has demonstrated the presence of this gene in S. aureus and CoNS isolates from milk, environmental, and human samples in mastitis cases in the Northeast region of Brazil (Pernambuco) [14]. Moreover, studies have highlighted that methicillin-resistant CoNS are globally recognized as a major cause of persistent infections in humans and animals, particularly S. epidermidis, S. haemolyticus, and Staphylococcus lugdunensis [54,55,56].

In the present study, no Staphylococcus spp. presented the mecC gene, although this gene has been the target of several studies in Brazil, there are only two studies with positive samples [15, 57]. In the Americas, one of the first reports of finding this gene in a case of bovine mastitis was in an isolate of Staphylococcus saprophyticus in Argentina [58]. However, in Brazil, there are only two reports on the occurrence of this gene in a Staphylococcus spp. isolate from a case of bovine mastitis; the first in the state of Pernambuco in the northeast of the country [15] and the second in the Southeast region of Brazil [57], both reports identified the mecC gene in S. aureus isolates.

Regarding the presence of vancomycin resistance genes (vanA and vanB), the presence of the vanB gene was observed in one isolate (1/110), a Staphylococcus chromogenes, and the absence of the vanA gene in all isolates. There are reports of the occurrence of the vanA (15/178) and vanB (1/178) genes in Brazil in Staphylococcus species isolated from the milk of goats with mastitis [59], but until the completion of this research, there are no reports of the occurrence of these genes in Staphylococcus species or in other species of bacteria isolated from bovine mastitis in Brazil. This finding is unprecedented in Brazil and is alarming for public health issues, configuring as the first record of this gene (vanB) in a bacteria isolated from bovine mastitis.

The findings of mecA and vanB positive bacteria in bovine mastitis samples if of major impact, especially for public health, since, consuming milk that has not been correctly processed may cause infection in humans. The presence of mecA and vanB carrying bacteria is indicative of possible horizontal transmission of resistant bacteria from humans to animals, since, methicillin and vancomycin are not used in the treatment of animal infection. Another worrying fact is the occurrence of both mecA and vanB genes in the same study since vancomycin is considered the first choice of antibiotics used in treatment against methicillin-resistant Staphylococcus (MRS) [16, 60,61,62,63,64,65,66].

In late years, strains have been identified carrying two or more resistance genes, showcasing that bacterium may present different resistance mechanisms [13, 52]. It was noted that some bacteria in our study also harbored more than one resistance gene (blaZ, tetL, and tetM). The mecA and vanB genes were not associated with other genes; however, it was observed that both were expressed in the phenotypical resistance test.

Some strains carrying these genes did not display similar behavior in the disk-diffusion test, since bacterial strains may carry a resistance gene but may not express this gene; the phenotypic expression of the gene depends on several factors such as environmental conditions and the genetic context [67].

Conclusion

The identification of all Staphylococcus species in the present study related to mastitis cases, as well as its characterization of the phenotypic and genotypic resistance profile of these isolates for some classes of antimicrobials, has a high impact on the dairy region, since it will allow for the elaboration of control measures against this disease. Also, it is worrying that the circulation of antimicrobial-resistant samples in dairy farming, considering that antimicrobials, such as vancomycin and methicillin, are not used in the treatment of animal infections in Brazil.