Background

Mastitis is a common disease in dairy herds, causing large declines in profitability and harmful effects on animal welfare [1]. It can be caused by a variety of microorganisms and Staphylococcus aureus is one of the most important pathogens isolated from the milk of cows with subclinical mastitis [2].

In 2015 whole genome sequencing (WGS) confirmed that isolates from several clonal S. aureus lineages were sufficiently divergent to be designated as separate coagulase-positive species, included in S. aureus complex (SAC). While S. schweitzeri appears to be predominantly associated with wildlife, being restricted to Sub-Saharan Africa [3], S. argenteus has been reported worldwide [4,5,6,7,8,9,10] and is predominantly human-associated, although their recovery from several animals already occurred [3]. This new species had already been described before S. aureus like CC75/ST 75 [11], with an evolutionary sequence very distant from other S. aureus strains. Nonetheless, S. argenteus presents more than 60 CC, such as CC2196 (35 STs), followed by CC1594 (12 STs), CC2198 (7 STs), CC75 (5 STs), and CC2793 (2 STs) [6, 12, 13].

S. argenteus has an average nucleotide identity of 87.4%, corresponding to a DNA-DNA hybridization value of 33.5% compared to S. aureus [12, 14]. S. argenteus shares most of the major virulence factors of S. aureus, such as staphylococcal enterotoxins (SEs) [15] Antimicrobial resistance genes such as mecA have also been found in S. argenteus [16, 17] indicating the presence of mobile genetic islands and horizontal gene transition.

Phenotypic differences could be observed after growth in chocolate agar, tryptone soya agar (TSA) and salt egg yolk agar (SEY), as only S. aureus has a carotenoid pigment named staphyloxanthin [7, 12, 18] The operon crtOPQMN is responsible for coding the enzymes CrtO, CrtP, CrtQ, CrtM and CrtN, which are essential for staphyloxanthin synthesis [19] and the failure of any of these enzymes leads to the production of white colonies by S. argenteus in most agar media [9].

Genotypically, both the specific thermonuclease gene, nuc and 16S rRNA genes, which are used for confirmation of the S. aureus species, have identical sequences compared to S. argenteus [6]. But there is a genetic difference that can be used as a discriminative method based on deletion of 60 amino acids in the sequence of S. aureus nonribosomal peptide synthetase (NRPS), which does not occur in S. argenteus. Zhang et al. [9] discovered this deletion by comparing open-reading frames of S. aureus NCTC 8325 (GenBank no. NC_007795) and S. argenteus MSHR 1132 T (GenBank no. NC_016941) genomes, using BLAST algorithm. These genomic comparisons confirmed the presence of the intact NRPS gene in S. argenteus. Otherwise, more elaborate techniques are needed to discriminate S. argenteus from S. aureus, such as multilocus sequence typing (MLST) [9, 20].

Reports of this new species and improvements in identification techniques are still recent and therefore strains previously classified as S. aureus may be reclassified as S. argenteus according to the new characteristics.

Thus, the objective of the study was to investigate the presence of S. argenteus in isolates previously classified as S. aureus, obtained from the milk of cows with clinical or subclinical mastitis in different regions of Brazil.

Results

Among the 856 strains (S1 Table) that were seeded onto the different agars, 185 (21.6%) had colonies that were white or colorless, but definitely not yellow. Interestingly, none of the 10 isolates from the state of Rio de Janeiro presented pigmentation, as indicated in Table 1. Among 185 non-pigmented strains, 60% of isolates (N = 111/185) presented the complete operon (Table 1). The crtQ gene was the most frequent, present in 157 out of 185 isolates (84.9%) (S2 Table).

Table 1 Presence of the operon crtOPQMN genes among non-pigmented S. aureus strains isolated from the milk of cows with clinical and subclinical mastitis, in different states in Brazil
Full size table

Considering all genes that make up the staphyloxanthin operon (crtO, crtP, crtQ, crtM, crtN), 74 out of 185 (40%) non-pigmented isolates did not present at least one of them, and crtO was the least frequent, according to Table 2.

Table 2 Genotypic profiles of operon crtOPQMN among non-pigmented S. aureus strains isolated from milk of cows with clinical or subclinical mastitis
Full size table

The nrps gene was searched only in strains lacking at least one of the staphyloxanthin operon genes. All 74 strains presented amplicons of approximately 160 bp, considered to be S. aureus according to Zhang et al. [9] as S. aureus has a deletion in this gene while S. argenteus has a product of approximately 340 bp. As all strains were confirmed to be S. aureus, no MLST was required.

Discussion

S. argenteus has been found in other regions of the world in human and animal isolates [3, 9, 21]. However, to date, this microorganism has not been reported in Brazil. According to Moradigaravand et al. [22], S. argenteus presents genes associated with plasmids and other mobile genetic elements previously observed with livestock-associated S. aureus, suggesting the need for further research into the presence of S. argenteus in dairy herds.

Among the 856 S. aureus this study, none were reclassified as S. argenteus, indicating the absence or low incidence of this new species in Brazilian herds. The Amazon is considered one of three main hot spots in the world, along with Australia and Southeast Asia [3] so it is important to research this new species in Brazil since the presence of a hot spot facilitates the spread of this bacteria.

Until now, few collections of S. aureus strains were retrospectively analyzed for the study of S. argenteus using culture media such as TSA [9] or chocolate agar [7]. SEY agar had been used only in isolation procedures for visualization of white colonies [18]. As a screening method, this phenotypic test, does not seem to be appropriate as we tested 856 isolates of bovine mastitis milk, with 185 (21.6%) non-pigmented and none of them were S. argenteus. Zhang et al. [9] tested 839 S. aureus isolated from food products, healthy humans and hospital infections in TSA, and 89 (10.6%) had white colonies, but only six of them (6.7%) were confirmed to be S. argenteus. Arguidin et al. [7] tested 1903 strains of S. aureus isolated from human infections in chocolate agar and found 73 (3.8%) non-pigmented isolates, but only three (4%) were S. argenteus.

Most (60%) of the non-pigmented S. aureus isolates from this study presented all operon crtOPQMN genes as did those in Zhang et al. [9] who detected the complete crtOPQMN operon in all their non-pigmented S. aureus. In those cases, the lack of pigmentation did not occur due to the absence of one of the genes so other factors must be involved in its expression or regulation.

Kitagawa et al. [23] investigated the absence of the crtM gene to confirm S. argenteus species but, as observed in this study, 46 strains confirmed as S. aureus also did not present the crtM gene, indicating that its absence is not an exclusive feature of S. argenteus.

Using the nrps gene [9], all non-pigmented isolates were confirmed as S. aureus, despite the absence of at least one gene in the staphyloxanthin operon. In this technique, the nrps gene product for S. argenteus has a molecular weight of approximately 340 bp while for S. aureus the molecular weight is reduced to approximately 160 bp. With this, the initial separation between the two species become quick and effective.

The results showed that screening tests using the lack of pigmentation in different culture media and the absence of the crtM gene did not show high discriminatory power in the differentiation between S. aureus and S. argenteus. On the other hand, screening using the nrps gene was efficient. MLST is considered the most reliable technique for confirming S. argenteus [9]. However, as it is a laborious technique, it needs more effective screening.

Conclusions

As far as we know, this is the first review of an S. aureus collection isolated from clinical and subclinical mastitis in Brazil, and S. argenteus is not yet a significant problem to be considered in cases of mastitis. Besides, the development of a better phenotypic screening test is needed before conducting molecular confirmation, which is more laborious and expensive.

Methods

Bacterial samples

A total of 856 strains previously identified as S. aureus were reevaluated. All strains were isolated from the milk of cows with clinical or subclinical mastitis, from 2010 to 2019, from five Brazilian states: São Paulo (n = 615), Minas Gerais (n = 130), Pernambuco (n = 57), Paraná (n = 44) and Rio de Janeiro (n = 10). The strains are part of the bacterial collection found in the Food Microbiology Laboratory of the Department of Microbiology and Immunology at the Institute of Biosciences of Paulista State University in Botucatu, São Paulo (Brazil). Previously, screening tests for S. aureus identification were gram stain, catalase and coagulase tests, and tests for clumping factor. Confirmation through molecular typing was performed by polymerase chain reaction (PCR) using the specific nuclease (nuc) gene [24]. This study was approved by the Committee on Ethics in Animal Use of UNESP (0136/2017).

Phenotypic and genotypic identification of S. argenteus

The initial screening was performed using SEY agar [18], chocolate agar (Oxoid, Basingstoke, UK) [7] and TSA agar (Oxoid, Basingstoke, UK) [9]. All isolates were incubated in Brain and Heart Infusion Broth (Oxoid, Basingstoke, UK) at 37 °C/24 h. After this period, an aliquot of each strain was seeded onto the different agars and incubated for 48 h at 37 °C. After this period, strains showing white colonies or without pigmentation (gray) were suspected of S. argenteus and selected for molecular analysis. The operon crtOPQMN genes (crtO, crtP, crtQ, crtM and crtN) were screened by PCR, according to Zhang et al. [9] in all nonpigmented isolates. The nrps gene [9] was tested only in isolates that did not have at least one of the crtOPQMN operon genes. S. aureus MW2, S. aureus WB49, S. aureus N315 and S. aureus COL were used as positive controls for the reactions.