Introduction, methods and results

Staphylococcus aureus remains a great challenge for bovine udder health in dairy herds, as its treatment and control when causing mastitis often fail due to the complex nature of its virulence factors. The accessory gene regulator (agr) locus, a well-known global regulatory system, is a master regulator of S. aureus virulence and affects the expression of numerous virulence factors involved in tissue colonization and invasion [1]. Giving the polymorphisms in agrB, agrC and agrD genes of this locus, four S. aureus agr specificity groups has been defined [2]—agrI, agrII, agrIII, and agrIV—and each of them has been associated with a specific type of mastitis [3]. In this regard, S. aureus agr group I is the most prevalent among isolates obtained from bovine subclinical mastitis [3] and is highly associated with intracellular survival in bovine mammary epithelial cells, favoring persistent intramammary infection [1].

We previously demonstrated that metabolites of bovine non-aureus staphylococci (NAS) have the capacity to modulate the virulence of a S. aureus agrI strain [4]. Therefore, we hypothesize that bovine NAS metabolites also modulate S. aureus invasion into bovine mammary gland epithelial cells. Thus, we aimed to evaluate whether supernatants (SN) from four well-studied bovine NAS isolates impact S. aureus adhesion to and internalization into bovine mammary epithelial cells and if so, to determine whether such effects may be associated with the agr system by comparing an agr-positive (agr+) with an agr-negative (agr) S. aureus strain.

We selected four bovine-associated NAS isolates based on our previous in vitro findings regarding the inhibitory effect of NAS against (methicillin-resistant) S. aureus and regarding the effect of NAS SN on the expression of S. aureus agr-related genes [4,5,6]. The S. chromogenes IM strain originating from a persistent intramammary infection [7] and the S. chromogenes TA strain isolated from a heifer’s teat apex (also referred as “C2” in [8]) were used as well as a S. simulans isolate (“SS10”) originating from a teat apex and a S. epidermidis isolate (“SE2”) originating from milk. The S. aureus strain 8325-4 (agr+ group I) [9] and its mutant strain 8325-4 Δagr (agr) [10], respectively, were included in this study as targets.

The cell-free SN were obtained from the cultures of the NAS isolates, as previously described [4]. Next, the β-galactosidase plate assay was performed to confirm the capacity of NAS SN to regulate the S. aureus agr-related virulence genes, with the assessment based on measurement of halo zone diameters (data not shown). As expected, our results substantiated the earlier findings that all NAS SN, except SN from the SE2 isolate, affect the expression of the S. aureus agr quorum sensing system [4]. The β-galactosidase plate assay was repeated (data not shown) confirming the previous findings that all NAS SN, except SN from the SE2 isolate, affect the expression of the S. aureus agr quorum sensing system [4]. Subsequently, cultures (2 × 104 CFU/mL) of both S. aureus strains were washed and resuspended in the different NAS SN. These cultures (NAS SN + S. aureus; n = 8) were allowed to interact for 2 h prior to challenging bovine mammary epithelial cells with the two S. aureus strains, adapted from and modified as follows [11].

A widely used clonal bovine mammary epithelial cell line (MAC-T) was cultured as previously described [12]. After 2 h of interaction between the NAS SN and the respective S. aureus strains, the cultures were centrifuged at 3220×g for 10 min and the S. aureus strains were recovered in MAC-T medium and added to the MAC-T cell culture (2 × 104 cells/well). As positive controls, both S. aureus strains without prior contact with NAS SN were also added to the cell culture (1:1 ratio bacteria/cells) and unchallenged MAC-T cells were kept as negative controls. The adhesion and internalization assays were assessed as described by Souza et al. [12]. After 4 h of incubation, the number of bacteria in the MAC-T medium, as well as adhesion and internalization were assessed, and used to calculate the percentage of adhered and internalized bacteria. Each co-culture condition was tested with six replicates.

Statistical analysis were performed using GraphPad Prism 9.4 (GraphPad Software, Inc., San Diego, USA). Adhesion and internalization data were transformed using the squared and z-scores values to obtain normally distributed outcomes variables, respectively. Then, the data were subjected to two-way ANOVA followed by Tukey’s test. It was analyzed whether the percentage of adhered and internalized S. aureus to and into MAC-T cells was influenced (1) by SN of all NAS combined as a group and (2) by SN of the four NAS isolates separately, respectively, as compared with the negative controls. The effect of the S. aureus strain (agr+ versus agr) was tested as well and the interaction with the NAS variables (NAS group/separate NAS isolates, respectively) was also included. Significance was set at P ≤ 0.05. Original median and interquartile range were preserved to improve the interpretation of the results.

The adhesion of the S. aureus agr+ strain onto the bovine mammary epithelial cells did not statistically differ from the S. aureus agr strain (P = 0.99; Figure 1A), whereas internalization into the MAC-T cells by the agr+ strain was more than tenfold higher compared with the agr strain (P < 0.0001; Figure 1C).

Figure 1
figure 1

Non-aureus staphylococci (NAS) supernatants affect Staphylococcus aureus adhesion to and internalization into MAC-T cells. Percentage of Staphylococcus (S.) aureus strains (agr+ and agr) adherent to (A and B) and internalized into (C and D) MAC-T cells with and without previous contact with cell-free supernatants from NAS isolates: S. chromogenes IM, S. chromogenes TA, S. simulans SS10 or S. epidermidis SE2. Results are presented as median ± interquartile range.

Prior contact with NAS SN (group) increased substantially S. aureus adherence to the bovine mammary epithelial cells (P = 0.0001 for the agr+ strain and P = 0.0009 for the agr strain; Figure 1A), an effect that was more outspoken for the agr+ compared with the agr− strain (P = 0.03; Figure 1A).

Individually, the SN of all different NAS isolates promoted the adhesion of S. aureus agr+ strain (Figure 1B), an effect that was also seen for S. aureus agr strain except for the S. chromogenes IM SN that did not significantly stimulate adhesion. The significant interaction term between the NAS SN effect and S. aureus agr system (P = 0.004) indeed underpins the different effects between the different NAS isolates and the two S. aureus strains.

Internalization in response to prior interaction with the NAS SN (group) was reduced fivefold for the S. aureus agr+ strain (P < 0.0001) whereas internalization was increased (P = 0.05) in the S. aureus agr strain.

All different NAS SN drastically inhibited the internalization of S. aureus agr+ to a comparable level, while S. chromogenes TA (P = 0.02) and S. epidermidis SE2 (P < 0.0001) actually promoted internalization of the S. aureus agr strain (Figure 1D). The significant interaction term between the NAS SN effect and S. aureus agr system (P < 0.0001) indeed substantiates different effects between the different NAS isolates and the two S. aureus strains.

Discussion

In this study, the capacity of the S. aureus agr+ strain (agr type I) to internalize into bovine MAC-T cells was confirmed, an effect that was almost absent in the agr mutant strain, supporting the hypotheses that the agr system is involved in tissue invasion. Previous findings showed that S. aureus agr type I, present in the S. aureus strain 8325-4 used here, can evade the immune system through internalization into bovine mammary epithelial cells significantly better than isolates carrying other agr types [1]. As well, it can persist in the host for a longer period without symptoms of apparent inflammation, also limiting the action of antimicrobials [1], which is very relevant for the control and treatment of bovine S. aureus mastitis. The agr system is responsible for the coordination of the transition to an invasive mode, which involves increased production of virulence factors and a decrease in surface proteins [13]. Therefore, the inhibition of the agr system could be the key to reduce S. aureus pathogenicity, antibiotic resistance, and biofilm formation.

In the light of the observations made in the present study, we suggest that the absence of the agr regulation, either by the inhibition of agr signaling by NAS SN (S. aureus agr+) or absence of a functional agr system (S. aureus agr), might be contributing (to some extent) to S. aureus adhesion to MAC-T cells. In this scenario, the increased adhesion of S. aureus after prior contact with NAS SN to MAC-T cells could potentially lead to a higher recruitment of neutrophils into the mammary gland, which is a well-known pivotal defense mechanism against invading pathogens, as bacterial adhesion to epithelial cells is a strong determinant for the production of neutrophil chemoattractants [14].

A promising finding of our study is that S. aureus agr+ internalization into MAC-T cells was markedly damped after contact with all NAS SN (Figure 1C), which is important for the pathogenesis. Interestingly, the udder-adapted S. chromogenes IM strain was able not just to reduce the internalization of S. aureus agr+, but also to hinder S. aureus agr internalization into the MAC-T cells (Figure 1D). These findings provide further evidence on the putative protective effect of the S. chromogenes IM strain in relation to bovine udder health [4, 5, 15,16,17].

Another intriguing finding is that S. aureus agr internalization into MAC-T cells increased after prior contact with SN from S. chromogenes TA and S. epidermidis SE2 (Figure 1D). Surprisingly, previous findings of our research group showed that the same NAS isolates were capable to promote biofilm formation of the S. aureus agr strain and to suppress biofilm dispersion of the same S. aureus strain [5]. Altogether, our results indicate that certain metabolites produced and secreted by NAS isolates, which are also present in the SN, may have the ability to influence the internalization of S. aureus via a mechanism unrelated to the agr system.

On one hand, the inactivation of the agr system results indeed in a greater potential for cell invasion [18]. On the other hand, the inhibition of agr signaling in S. aureus agr+ strain by NAS SN most probably triggered a shift of the expression of exoproteins such as spa [4, 18] that decreases cell internalization capacity. However, intrinsic factors from both S. aureus strains and SN from NAS isolates may have potentially contributed to the differences in the degrees of internalization observed in our study. Consistent with the unique behavior exhibited by NAS strains, prior research has revealed significant variations in virulence [19], potential beneficial properties [4, 5, 8, 16], and host-interaction [12, 20, 21] among and within bovine-associated NAS species. Hence, further studies are required to explore and identify metabolites produced and secreted by distinct NAS, their nature of action, and their potential use for combating infections, including bovine intramammary infection.

We conclude that the mechanism behind the dichotomous behavior of S. aureus adhesion to and internalization into bovine mammary epithelial cells by NAS SN is NAS isolate-dependent, and this effect is likely reliant on the S. aureus agr system although we cannot dismiss the potential involvement of unmeasured S. aureus traits and particular NAS metabolites. Still, as internalization of S. aureus into bovine mammary epithelial cells plays a key role in immune evasion [12] and that bacterial adherence to epithelial cells is as a strong determinant for neutrophil chemoattractant production [14], our findings open new perspectives to explore innovative anti-virulence strategies targeting treatment and control of bovine S. aureus mastitis.