Lineage associated expression of virulence traits in bovine-adapted Staphylococcus aureus

Veterinary Microbiology 189 (2016) 24–31

Contents lists available at ScienceDirect

Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic

Lineage associated expression of virulence traits in bovine-adapted Staphylococcus aureus Kathleen E. Budda,b , Jennifer Mitchellb , Orla M. Keanea,* a b

Animal & Bioscience Department, AGRIC, Teagasc, Grange, Dunsany, Co. Meath, Ireland School of Veterinary Medicine, University College Dublin, Belfield, Dublin 4, Ireland

A R T I C L E I N F O

Article history: Received 25 January 2016 Received in revised form 24 March 2016 Accepted 18 April 2016 Keywords: Staphylococcus aureus Mastitis Biofilm Immune response Infection Internalisation

A B S T R A C T

Bovine mastitis is the most costly disease to the dairy industry worldwide with Staphylococcus aureus commonly associated with intramammary infections that are persistent and refractory to treatment. The strains of S. aureus that cause mastitis predominantly belong to a number of well-described bovineadapted lineages. The objective of this study was to determine if a variety of potential virulence traits were associated with lineage. Bovine-adapted S. aureus isolates (n = 120), belonging to lineages CC97, CC151 and ST136, were tested for their ability to adhere to and internalise within cultured bovine mammary epithelial cells (bMEC), to bind bovine fibronectin, to form a biofilm in TSB, TSB + 1% glucose and TSB + 4% NaCl, and to induce an immune response from bMEC. There were no significant differences between the lineages in ability to adhere to or internalise within bMEC although there were significant differences between individual isolates. For lineages CC97 and ST136, mammalian cell adherence was correlated with the ability to bind bovine fibronectin, however isolates from CC151 could not bind bovine fibronectin in vitro, but adhered to bMEC in a fibronectin-independent manner. There were significant differences between the lineages in ability to form a biofilm in all three growth media with ST136 forming the strongest biofilm while CC151 formed the weakest biofilm. Lineages also differed in their ability to elicit an immune response from bMEC with CC97 eliciting a stronger immune response than CC151 and ST136. These data indicate the potential for both lineage and strain-specific virulence and a strainspecific response to infection in vivo and caution against extrapolating an effect from a single strain of S. aureus to draw conclusions regarding virulence or the host response to infection in unrelated lineages. ã 2016 Elsevier B.V. All rights reserved.

1. Introduction Staphylococcus aureus is among the most common pathogens associated with bovine intramammary infection (IMI), a disease of global importance with negative consequences for animal health and welfare, and milk quality (Keane et al., 2013; Peton and Le Loir, 2014; Wickstrom et al., 2009). S. aureus is associated with both clinical and more commonly sub-clinical mastitis, both of which frequently result in persistent and recurrent infections with a low cure rate after antibiotic therapy (Schukken et al., 2011). The mechanism of persistence of S. aureus is still not fully elucidated, although a number of strategies have been postulated. One mechanism used by S. aureus to evade the immune system is internalisation into host cells. S. aureus can invade a variety of nonprofessional phagocytes (Haggar et al., 2003) and persistent

* Corresponding author. E-mail address: [email protected] (O.M. Keane). http://dx.doi.org/10.1016/j.vetmic.2016.04.013 0378-1135/ã 2016 Elsevier B.V. All rights reserved.

infections may be related to the ability of S. aureus to invade and survive within certain types of host cells. It is well established that S. aureus can adhere to and internalise into mammary gland epithelial cells (MEC). The bacteria may also evade phagocytosis by persisting in the form of metabolically inactive small colony variants (Atalla et al., 2008). The best understood mechanism for S. aureus internalisation is provided by fibronectin binding proteins (FnBPs). Fibronectin acts as a bridging molecule which binds Nterminally to S. aureus FnBPs and via an RGD motif to the host cell integrin a5b1 and signals bacterial uptake (Fowler et al., 2000). Chronic, persistent and antibiotic-refractory S. aureus infections have also been associated with growth of the bacteria in a biofilm (Dunne, 2002). Biofilms are a community of bacterial cells arranged in a structured manner, enclosed in a self-produced polymeric matrix and adherent to an inert or living surface (Costerton et al., 1999). S. aureus can produce a number of molecules which facilitate biofilm formation including an extracellular polysaccharide adhesin, extracellular DNA and a variety of proteins such as the biofilm associated protein (Bap). Biofilm-

K.E. Budd et al. / Veterinary Microbiology 189 (2016) 24–31

associated bacteria show an innate resistance to antibiotics and clearance by host defense mechanisms (Melchior et al., 2006) and a bap+ S. aureus strain had a significantly enhanced ability to colonize and persist in the ovine mammary gland compared to a bap
strain (Cucarella et al., 2004). The innate immune system is the first line of defense against an invading pathogen. Pattern recognition receptors on the surface of host cells recognize non-specific pathogen associated molecular patterns (PAMPs) such as lipopolysaccharide or lipoteichoic acid. This induces a signaling cascade which results in the production of cytokines, chemokines and the recruitment of somatic cells to the mammary gland (Brenaut et al., 2014). MEC are the most abundant cell type to initially encounter intramammary pathogens and are an important source of pro-inflammatory cytokines and chemokines that attract neutrophils early in infection, with professional immune cells such as dendritic cells and macrophages also essential for immune surveillance. The ability to mount a rapid and effective immune response is critical to resolving a mastitis infection (Thompson-Crispi et al., 2014). However, S. aureus has acquired a variety of immune evasion strategies to attenuate the host immune response and facilitate pathogen persistence (Foster, 2005; Vrieling et al., 2015). Particular lineages of bovine-adapted S. aureus are responsible for the majority of cases of bovine infections worldwide (Smith et al., 2005). There is substantial genetic variation between lineages, with gene and allele content of a variety of virulence factors lineage-specific (Budd et al., 2015; McCarthy and Lindsay, 2010) suggesting lineages may differ in their molecular mechanisms of pathogenicity. An improved understanding of the association between lineage and virulence traits would be valuable for the development of S. aureus IMI control and treatment strategies. A previous study found that the lineages associated with clinical mastitis in Ireland were CC97, CC151 and ST136 (Budd et al., 2015). The objective of this study was to determine if lineage was a predictor of a number of virulence attributes, including the ability to adhere to and internalize within MEC, form a biofilm and modulate the host response to infection. 2. Materials and methods 2.1. Bacterial isolates and growth conditions All S. aureus bovine-adapted isolates used in this study were recovered from milk samples taken from cows presenting clinical mastitis between February 2010 and February 2011 from 26 farms in Ireland. Sample collection, bacterial isolation and identification methods and isolate genotyping have been described previously (Budd et al., 2015; Keane et al., 2013). A brief description of the lineages used in the study can be found in Table 1. Reference strains SH1000, SH1000DfnbABDclfAB and Cowan (ATCC 12598) were a kind gift from Prof Tim Foster (TCD). All strains were cultured in trypticase soy broth (TSB) or trypticase soy agar (TSA) at 37  C unless otherwise stated.

25

2.2. Adherence and internalisation Quantification of S. aureus adherence to and internalisation within mammalian cells by flow cytometry was previously described (Trouillet et al., 2011). The bovine mammary epithelial cell line, MAC-T, was grown in Dulbecco’s modified eagle medium (DMEM) supplemented with 10% (v/v) fetal bovine serum (FBS) at 38  C with 5% (v/v) CO2. Cells were disassociated from the flask using Trypsin-EDTA (0.25%), and centrifuged at 118g for 8 min. The supernatant was removed and the cells gently resuspended in 10 mL of DMEM + FBS. The centrifugation and resuspension steps were repeated twice more. All bacterial isolates were grown in 5 mL TSB for 16 h at 37  C. The bacterial cells were washed by centrifugation at 10,000g for 5 min, supernatant removed and resuspended in 5 mL sterile phosphate buffered saline (PBS). The optical density (OD) of the suspension was adjusted to an OD600nm of 1 using sterile PBS. Bacterial cells were stained with Calcein AM (Merck Millipore, UK) for 30 min at a 1:500 dilution. The cells were subsequently washed 3 times with sterile PBS. MAC-T cells were stained for 30 min with a 1:500 dilution of 40 ,6-Diamidino-2phenylindole dihydrochloride (DAPI) (Sigma Aldrich UK) and washed with sterile DMEM + FBS. Stained bacterial and MAC-T cells were counted on the Attune flow cytometer (Life technologies, Germany) before co-incubation at a multiplicity of infection (MOI) of 10:1 for 3 h at 38  C with 5% CO2. This MOI was chosen as it has previously been reported to lie in the mid-range of values for which there is a linear relationship between internalisation and MOI (Bayles et al., 1998). Cells were then washed 3 times with PBS, following which the murine primary antibody mAb55 (Hycult biotech, USA) was incubated with the cell/bacteria suspension at 1:150 for 30 min. Cells were subsequently washed three times with sterile DMEM + FBS. To detect bound mAb55, Qdot 655 donkey anti-mouse IgG was diluted 1:100 in the cell/bacteria suspension and incubated for 30 min before washing with DMEM + FBS three times. Aliquots of the suspension were added to the wells of a 96 well round bottom plate (Starstedt, Germany) and mammalian cells counted on the Attune flow cytometer. The number of MAC-T cells with bacteria internalized was calculated by subtracting the number of cells stained with mAb 55 (adhered only) from the number stained with calcein AM (adhered and internalized). Results are the mean of triplicate experiments. 2.3. Fibronectin binding Fibronectin-binding was determined as described previously (Zapotoczna et al., 2013). Flat bottom 96-well ELISA plates (Starstedt, Germany) were coated with 100 mL of 0.625 mg/mL bovine fibronectin (Calbiochem, Merck Chemicals, Germany) overnight at 4  C. Wells were then washed 3 times with PBS, blocked with 5% BSA for 2 h at 37  C and again washed 3 times with PBS. Bacterial strains were grown overnight in TSB at 37  C, diluted 1:200 in pre-warmed TSB and grown to an OD600nm of 0.3, washed and resuspended to an OD600nm of 0.45 and added to triplicate

Table 1 Genotypic and phenotypic characteristics of the bovine-adapted S. aureus lineages. Lineage

Clonal Complex

Number of isolates

Key genotypic characteristics

CC71 tCC97 ST136 CC151

CC97 CC97 ST136 CC151

42 22 14 42

cap8 cap5 cap5 cap8

agrI bap
ica
bla+ agrI bap
ica+ bla+ agrIII bap
ica+ bla
agrII bap
ica+ bla


Key phenotypic characteristics FnBP+ SpA+ FnBP+ SpA+ FnBP+ SpA+ FnBP
SpA


26

K.E. Budd et al. / Veterinary Microbiology 189 (2016) 24–31

wells of the plate. The plate was covered and incubated statically for 2 h at 37  C. Non-adherent cells were removed by washing three times with PBS. Formaldehyde, 100 mL of 25% (v/v), was added to each well and left at room temperature for 20 min to fix the cells. The plate was subsequently washed 3 times with PBS and 100 mL of crystal violet (Sigma Aldrich, UK) was added to each well for 1 min. The plate was again washed with PBS until all unbound stain was removed, after which 100 mL 5% (v/v) Acetic Acid (Sigma Aldrich, UK) was added to the wells to solubilise the crystal violet and absorbance at 570 nm was determined. S. aureus SH1000 and S. aureus SH1000DfnbABDclfAB were used as positive and negative controls between and within plates.

each isolate. A transformation was applied to biofilm measurements (ln(X  100 + 1)) and ELISA measurements (sqrt(X)) to stabilize the variance. Data were analysed using the Proc MIXED procedure of SAS (v9.3). Lineage was the only effect used in the model, with the model for fibronectin and IL-1b accounting for the unequal variance between lineages. Multiple testing was adjusted for using the Tukey correction. In the case of adherence and internalisation isolates were also tested in order to determine if there were significant differences between them independent of lineage. In this instance triplicate phenotype measurements were not averaged and isolate was the only effect used in the model. A probability level of P < 0.05 was considered significant.

2.4. Biofilm formation

3. Results

Biofilm formation was determined as previously described (Merritt et al., 2005) with some minor modifications. Briefly, isolates were grown overnight in 5 mL TSB at 37  C and diluted 1:100 in TSB, TSB + 1% glucose or TSB + 4% NaCl. Diluted bacteria (100 mL) were added to sterile non-tissue culture treated 96 well round bottom polystyrene plates (Starstedt, Germany). The plates were covered and incubated statically at 37  C for 24 h. Planktonic bacteria were removed and wells washed by submerging in water. Crystal violet solution (125 mL) (Sigma Aldrich, UK) was added to each well and left at room temperature for 10 min. The plate was then washed twice in water. Plates were allowed to dry before 200 mL of 30% (w/v) acetic acid (Sigma Aldrich, UK) was added to each well, the dye allowed to solubilize, 125 mL removed to a new 96 well flat-bottom plate and optical density measured at a wavelength of 570 nm. S. aureus strain Cowan was used as an interplate control. Experiments were carried out in triplicate and each independent experiment consisted of 4 technical replicates.

In order to determine if bovine-adapted S. aureus lineages differed in virulence attributes, a panel of bovine mastitisassociated S. aureus isolates was examined for an array of phenotypic characteristics. The isolates consisted of the bovineadapted lineages CC97, CC151 and ST136. A genomic rearrangement has occurred within CC97 near the origin of replication which results in a sub-group of isolates related to ST71 that differ substantially over >300 kb including the loss of >30 kb of DNA (Budd et al., 2015). As members of this sub-group (CC71) differ from typical CC97 (tCC97) in a number of virulence genes, including the polysaccharide intercellular adhesin (ica) operon, collagen-binding protein (cna) and capsule type, they were treated as a separate lineage for analysis purposes resulting in four lineages, CC71 (n = 42), tCC97 (n = 22), CC151 (n = 42) and ST136 (n = 14).

2.5. Host immune response

The ability of the S. aureus isolates to adhere to and internalize within MAC-T bovine mammary epithelial cells (bMEC) was quantified by flow cytometry and individual isolate results are available in Supplementary Table 1. There was a large amount of variation among the isolates with the percentage of MAC-T cells with S. aureus adhered varying between 19% and 98% while the percentage of MAC-T with S. aureus internalized varied between 0.4% and 60%. There were no differences between the lineages in either adherence or internalisation (P > 0.05; Fig. 1A and B). However, there were significant differences between individual isolates and overall isolate was a significant source of variation in both adherence (P < 0.0001) and internalisation (P < 0.0001) reflecting substantial isolate to isolate variation independent of lineage. Isolates that were significantly different from each other are indicated in Supplementary Table 2.

The host immune response to 47 randomly selected isolates was determined. These were tCC97 (n = 10), CC71 (n = 15), CC151 (n = 17) and ST136 (n = 5). MAC-T cells were infected as described above but infections were continued for 12 h. This timepoint was chosen as a preliminary infection demonstrated that the inflammation marker serum amyloid A was not detected until a minimum of 8 h post-infection. Supernatants were collected, centrifuged at 2000g for 10 min at 4  C and frozen at
20  C. The levels of IL-1b, IL-6, IL-8 (Thermo Scientific, USA) and TNFa (R&D systems, USA) were assessed by enzyme-linked immunosorbent assay according to the manufacturer’s protocol. Briefly, capture antibody was diluted in PBS and 100 mL used to coat each well of a 96-well ELISA plate. To duplicate wells of the plate, 100 mL of sample or standard was added, the plate covered and incubated for 1 h at room temperature with moderate shaking. The plate was then washed with wash buffer (PBS + 1% Tween) and biotinlabelled detection antibody diluted 1:100 was added to each well. The plate was incubated and washed as previously described. The streptavidin horse radish peroxidase conjugate was diluted 1:400 and added to each well, covered and incubated for 30 min at room temperature. The plate was washed 3 times, 100 mL TMB added to each well and incubated in the dark for 20 min at room temperature. The reaction was stopped by adding 100 mL stop solution. The absorbance was measured at 450 nm and 550 nm. Infections were performed in triplicate and values below the limit of detection were treated as zero. 2.6. Statistical analysis In order to determine if phenotypes differed between the lineages, triplicate phenotype measurements were averaged for

3.1. Adherence and invasion

3.2. Fibronectin binding The ability of the isolates to bind to the extracellular matrix protein fibronectin was also quantified (Supplementary Table 1). There was a significant difference between the lineages in their ability to bind to bovine fibronectin, with CC151 binding fibronectin at a significantly lower level than the other three lineages (P < 0.0001). There was no difference among CC71, tCC97 or ST136 in their ability to bind bovine fibronectin (Fig. 1C). Fibronectin binding is a key determinant of the ability of S. aureus to adhere to mammalian cells and fibronectin binding ability was positively correlated with adherence for the lineages CC71, tCC97 and ST136 (Fig. 1D; R = 0.67; P < 0.0001). Interestingly, ability to bind fibronectin was not correlated with mammary cell adherence for lineage CC151. While CC151 showed the same range of values for adherence as the other lineages, it did not bind the cells in a fibronectin-dependent manner (Fig. 1D).

K.E. Budd et al. / Veterinary Microbiology 189 (2016) 24–31

3.3. Biofilm formation The ability of each isolate to form a biofilm in TSB, TSB + 1% glucose and TSB + 4% NaCl was quantified (Supplementary Table 1). Lineage ST136 formed the strongest biofilm in all 3 media tested while CC151 formed the weakest biofilm in all 3 media. There were significant differences in biofilm formation between the lineages in each of the three growth media (Fig. 2; P < 0.0001). In TSB and TSB + 1% glucose all lineages were significantly different from each other with the exception of the closely related groups CC71 and tCC97. In TSB + 4% NaCl all lineages were significantly different from each other with the exception of tCC97 and CC151. Growth media also had a significant effect on the ability to form a biofilm with glucose inducing biofilm formation (P < 0.0001) in all lineages while NaCl repressed biofilm formation in tCC97 (P < 0.0001) and ST136 (P < 0.05). However, the effect of growth media was not always consistent among all isolates within a lineage as evidenced by the response to NaCl among isolates belonging to lineage CC71. Within this lineage 2 sub-groups were clearly discernible i) sub-group ST3173 (n = 10) which induced biofilm production in response to high osmolarity ii) sub-group

27

ST71 (n = 27) which repressed biofilm production in conditions of high osmolarity (Supplementary Fig. 1). 3.4. Host immune response The ability of a subset of isolates from the different lineages (n = 47) to induce an immune response from bMEC was quantified at 12 h post-infection. This time period was chosen to ensure sufficient time for a cytokine response to be generated by the host cells. The pro-inflammatory response was determined by quantitative sampling of the levels of IL-1b, IL-6, IL-8 and TNFa and results are shown in Supplementary Table 1 and Fig. 3. The concentration of IL-8 and TNFa induced by S. aureus infection was very low and in many cases below the limit of detection for a number of samples and so the effect of lineage on IL-8 or TNFa production could not be determined. The highest concentration of IL-1b was induced by tCC97 followed by CC71. These lineages also induced the highest concentrations of IL-6. There was a significant difference between the lineages in IL-1b production (P = 0.0004, Fig. 3A) with infection with ST136 resulting in less IL-1b than tCC97 and CC71. CC151 also induced significantly less IL-1b than

Fig. 1. Phenotypic characteristics of bovine adapted S. aureus lineages. Mean (s.e.) adherence (A), internalisation (B) and fibronectin binding (C) for the bovine-adapted lineages CC71 (n = 42), tCC97 (n = 22), ST136 (n = 14) and CC151 (n = 42). D shows a scatterplot of mean fibronectin binding plotted against mean mammalian cell adherence for each isolate. Regression lines for each lineage are also shown. ****P < 0.0001.

28

K.E. Budd et al. / Veterinary Microbiology 189 (2016) 24–31

tCC97. There was also a significant difference between the lineages in IL-6 induction (P = 0.0004, Fig. 3B). In this instance CC151 induced significantly less IL-6 than both tCC97 and CC71. 4. Discussion Adherence is an essential step required for S. aureus colonisation of the mammary gland and is mediated by a variety of surface bound proteins that interact with host ligands. Genetic variation in S. aureus surface protein-encoding genes has been reported to be lineage-specific, although many domain variants were conserved across unrelated lineages (Budd et al., 2015; McCarthy and Lindsay, 2010). Notably, CC151 isolates do not express a number of surface proteins including Protein A, FnBPA and FnBPB and it has been suggested that CC151 may be less virulent than other bovineadapted lineages (Stutz et al., 2011). FnBPs are key determinants of the ability of S. aureus to adhere to and invade mammalian cells. A strain lacking FnBP expression was reported to reduce MAC-T cell adherence by 40% and invasion by 95% compared with the isogenic parent strain (Dziewanowska et al., 1999). FnBPs have also been reported to be important virulence factors under milk flow in a mouse mastitis model (Brouillette et al., 2003). Therefore, CC151 may be expected to be deficient in mammalian cell adherence. Indeed it has been reported that isolates from CC151 displayed lower adhesion to primary bMEC than other isolates although this was based on only 3 isolates (Zbinden et al., 2014). We found that although CC151 lacked the ability to bind bovine fibronectin in vitro there was no difference among the lineages in bMEC adherence or internalisation, although there were significant differences between individual isolates. For the lineages CC71, tCC97 and ST136, adherence was positively correlated with fibronectin-binding, however, there was no correlation between fibronectin-binding and adherence for CC151. Despite their inability to bind bovine fibronectin, CC151 isolates displayed the same range of mammalian cells adherence as the other lineages indicating that this lineage utilizes

an alternative mechanism for bMEC adherence. A number of FnBPindependent mechanisms which facilitate S. aureus adherence to and internalisation within non-professional phagocytes have been described including internalisation mediated by Eap (extracellular adherence protein), IsdB (iron-regulated surface determinant B), and Atl (major autolysin); which interact with plasma proteins, mammalian integrins and Hsc70 respectively (Bur et al., 2013; Hirschhausen et al., 2010; Zapotoczna et al., 2013). Lpl lipoproteins encoded on the vSaa genomic island have also recently been reported to contribute to invasion (Nguyen et al., 2015) and such mechanism may account for the ability of CC151 to successfully invade bMEC. Biofilm formation has been associated with S. aureus isolates derived from bovine milk compared with teat skin or milking machine liners (Fox et al., 2005). Additionally, biofilm formation has been reported to facilitate adherence to ovine mammary epithelial cells and colonization of the ovine mammary gland (Cucarella et al., 2004). The biofilm matrix can be composed of a variety of molecules including polysaccharides, proteins and extracellular DNA. The polysaccharide biofilm matrix primarily consists of poly-N-acetylglucosamine (PNAG), the product of the intercellular adhesion (ica) locus while the proteinaceous biofilm can include the products of many genes including bap,fnb, sasG, clfB and sdrC. All isolates in this study were negative for the bap gene, previously associated with the ability to cause persistent mastitis (Cucarella et al., 2004) while isolates from CC71 lack the ica operon and CC151 does not express FnBPs. This suggested that lineages may differ in their ability to form a biofilm and the mechanism by which they form a biofilm; and indeed lineage-associated differences were observed. Various environmental factors have been reported to modulate biofilm formation (Dobinsky et al., 2003) and so the effect of glucose and salt on biofilm expression was tested. Glucose has been demonstrated to stimulate PNAG production (Cue et al., 2012) although cases of PNAG-independent glucosemediated biofilm formation have also been reported (O’Neill et al., 2007). For the isolates in this study, glucose increased biofilm

Fig. 2. Biofilm formation in bovine adapted S. aureus lineages. Back-transformed mean (with 95% C.I.) biofilm formation in TSB (black bars), TSB + 1% glucose (red bars) and TSB + 4% NaCl (green bars) for the bovine-adapted lineages CC71 (n = 42), tCC97 (n = 22), ST136 (n = 14) and CC151 (n = 42). Within each growth medium different superscripts denote significant differences between lineages. Identical superscripts denote no significant difference.

K.E. Budd et al. / Veterinary Microbiology 189 (2016) 24–31

29

Fig. 3. Cytokine production by bovine mammary epithelial cells in response to challenge with bovine adapted S. aureus. Back-transformed mean (with 95% C.I.) IL-1b (A) and IL-6 (B) produced by bovine mammary epithelial cells in response to infection with the bovine-adapted lineages CC71 (n = 14), tCC97 (n = 10), ST136 (n = 4) and CC151 (n = 17). *P < 0.05, **P < 0.01, ***P < 0.001.

production in all 4 lineages, including the ica-deficient CC71, implicating an ica-independent mechanism in the response to glucose, at least in this lineage. Despite CC71 lacking the ica operon, there was no difference between CC71 and tCC97 in biofilm formation in TSB or TSB + glucose. The apparent difference in TSB + NaCl was due to the heterogeneity of the response to NaCl among CC71 isolates; ST3173 induced biofilm production in response to NaCl while ST71 and other CC71 isolates repressed biofilm production. This demonstrates the limitation of lineage in capturing the full extent of variation between isolates. In human-associated S. aureus strains both glucose and salt have been generally associated with increased biofilm production in vitro (Lim et al., 2004; Rachid et al., 2000). In contrast to human-associated isolates, salt did not

promote biofilm production in any of the bovine-adapted lineages, although it did in ST3173. This suggests differences in the environmental regulation of biofilm production between human and the majority of bovine-adapted strains. It has previously been reported that strains showing higher levels of invasion of MAC-T cells were moderate to strong biofilm producers in TSB + 0.25% glucose (Bardiau et al., 2014). In our study, there was no relationship between biofilm production and ability to internalize within bMEC although there were a number of differences between our study and that of Bardiau et al., including a different MOI and different co-incubation times in the invasion assays and different glucose concentrations in the biofilm assays. An effective immune response is essential for intramammary pathogen clearance. S. aureus is considered to circumvent the host

30

K.E. Budd et al. / Veterinary Microbiology 189 (2016) 24–31

immune response as it generally induces a more muted immune response than E. coli with lower somatic cell count (SCC) and often results in sub-clinical or persistent infections that are not cleared from the mammary gland (Wellnitz and Bruckmaier, 2012). Mammary gland invasion by S. aureus results in contact between the infecting pathogen and both milk somatic cells and MEC. Pathogen recognition by MEC is crucial in the initial response against intramammary infections and results in the upregulation of key response genes. This study used MAC-T cells which have been validated as a suitable model of primary bMEC with a similar pattern and kinetics of immune gene induction in response to infection, albeit a weaker response (Gunther et al., 2016). Some of the key immune regulators are the pro-inflammatory cytokines IL1b, IL-6 and TNFa and the chemokine IL-8. IL-1, IL-6 and TNFa are key mediators of the local and systemic immune response and regulate the expression of a variety of genes including other cytokines; while IL-8 is a key molecule involved in the attraction of neutrophils to the udder (Wellnitz and Bruckmaier, 2012). It has previously been reported that infection with S. aureus results in the upregulation of IL-1b, IL-6 and IL-8 but not TNFa (Breyne et al., 2015; Wellnitz and Bruckmaier, 2012), although there is some disagreement among studies. In this study only IL-1 and IL-6 were expressed at appreciable levels 12 h post-infection of bMEC; with IL-8 and TNFa expression very low or below the limit of detection. There was a significant difference between the lineages with CC151 and ST136 inducing significantly less IL-1 from bMEC than tCC97 while CC151 induced less IL-6 than both tCC97 and CC71. Although, CC151 did not appear to induce a strong immune response compared with CC97, RF122 (CC151) has been reported to reproducibly induce severe mastitis while Newbould 305 (CC97) reproducibly induces mild and chronic mastitis (Peton et al., 2014). Whether this can be generalized to other strains from these lineages remains to be determined. It has previously been reported that the specific-infecting strain of S. aureus influences cytokine gene expression in bovine mammary epithelial cells in vitro (Lahouassa et al., 2007). However, the presence of fetal calf serum (FCS) or milk proteins in the growth medium has been reported to depress the immune response to S. aureus in MEC and to abrogate the S. aureus strain-specific immune response (Bauer et al., 2015). In this study, growth medium was supplemented with FCS and it would be interesting to examine if a more pronounced strainspecific host response could be observed under different culture conditions. In agreement with our study, the study by Zbinden et al. (2014) found lower expression of immune response genes in SLAT- strains (CC151) compared with other bovine-adapted strains. 5. Conclusion This study demonstrated significant differences in vitro among isolates and between lineages in virulence-associated traits such as epithelial cell binding and internalisation, fibronectin-binding, biofilm formation and ability to elicit immune response from bMEC. These data indicate the potential for a lineage-specific or strain-specific response to infection in vivo which merits further investigation. As mastitis diagnosis and the selection of mastitisresistant animals are primarily based on their inflammatory response (ie SCC) the ability of the infecting strain to modulate the immune response may influence not just the ultimate outcome of infection but treatment and management decisions. Knowledge of the phenotypic characteristics of the infecting strain is of crucial importance in the study of mastitis pathogenesis and the subsequent host response and when comparing studies of bovine mastitis. Importantly, caution may need to be exercised regarding the generalization of effects of a single strain or a strain from an unrelated lineage. However, further work is required to determine

the relevance of the in vitro phenotypes studied here to the in vivo infection process. Conflict of interest All authors declare that they have no conflict of interest. Acknowledgements The authors gratefully acknowledge the assistance of Margaret Murray (Teagasc) with the ELISA assays, Joan Geoghegan (TCD) and Tim Foster (TCD) with the fibronectin-binding assays and biofilm assays and Jim Grant (Teagasc) for statistical advice. This study was funded by Teagasc grant number 6082. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. vetmic.2016.04.013. References Atalla, H., Gyles, C., Jacob, C.L., Moisan, H., Malouin, F., Mallard, B., 2008. Characterization of a Staphylococcus aureus small colony variant (SCV) associated with persistent bovine mastitis. Foodborne Pathog. Dis. 5, 785–799. Bardiau, M., Detilleux, J., Farnir, F., Mainil, J.G., Ote, I., 2014. Associations between properties linked with persistence in a collection of Staphylococcus aureus isolates from bovine mastitis. Vet. Microbiol. 169, 74–79. Bauer, I., Gunther, J., Wheeler, T.T., Engelmann, S., Seyfert, H.M., 2015. Extracellular milieu grossly alters pathogen-specific immune response of mammary epithelial cells. BMC Vet. Res. 11, 172. Bayles, K.W., Wesson, C.A., Liou, L.E., Fox, L.K., Bohach, G.A., Trumble, W.R., 1998. Intracellular Staphylococcus aureus escapes the endosome and induces apoptosis in epithelial cells. Infect. Immun. 66, 336–342. Brenaut, P., Lefevre, L., Rau, A., Laloe, D., Pisoni, G., Moroni, P., Bevilacqua, C., Martin, P., 2014. Contribution of mammary epithelial cells to the immune response during early stages of a bacterial infection to Staphylococcus aureus. Vet. Res. 45, 16. Breyne, K., De Vliegher, S., De Visscher, A., Piepers, S., Meyer, E., 2015. Technical note: a pilot study using a mouse mastitis model to study differences between bovine associated coagulase-negative staphylococci. J. Dairy Sci. 98, 1090–1100. Brouillette, E., Talbot, B.G., Malouin, F., 2003. The fibronectin-binding proteins of Staphylococcus aureus may promote mammary gland colonization in a lactating mouse model of mastitis. Infect. Immun. 71, 2292–2295. Budd, K.E., McCoy, F., Monecke, S., Cormican, P., Mitchell, J., Keane, O.M., 2015. Extensive genomic diversity among bovine-Adapted Staphylococcus aureus: evidence for a genomic rearrangement within CC97. PLoS One 10, e0134592. Bur, S., Preissner, K.T., Herrmann, M., Bischoff, M., 2013. The Staphylococcus aureus extracellular adherence protein promotes bacterial internalization by keratinocytes independent of fibronectin-binding proteins. J. Invest. Dermatol. 133, 2004–2012. Costerton, J.W., Stewart, P.S., Greenberg, E.P., 1999. Bacterial biofilms: a common cause of persistent infections. Science 284, 1318–1322. Cucarella, C., Tormo, M.A., Ubeda, C., Trotonda, M.P., Monzon, M., Peris, C., Amorena, B., Lasa, I., Penades, J.R., 2004. Role of biofilm-associated protein bap in the pathogenesis of bovine Staphylococcus aureus. Infect. Immun. 72, 2177–2185. Cue, D., Lei, M.G., Lee, C.Y., 2012. Genetic regulation of the intercellular adhesion locus in staphylococci. Front. Cell. Infect. Microbiol. 2, 38. Dobinsky, S., Kiel, K., Rohde, H., Bartscht, K., Knobloch, J.K., Horstkotte, M.A., Mack, D., 2003. Glucose-related dissociation between icaADBC transcription and biofilm expression by Staphylococcus epidermidis: evidence for an additional factor required for polysaccharide intercellular adhesin synthesis. J. Bacteriol. 185, 2879–2886. Dunne Jr., W.M., 2002. Bacterial adhesion: seen any good biofilms lately? Clin. Microbiol. Rev. 15, 155–166. Dziewanowska, K., Patti, J.M., Deobald, C.F., Bayles, K.W., Trumble, W.R., Bohach, G. A., 1999. Fibronectin binding protein and host cell tyrosine kinase are required for internalization of Staphylococcus aureus by epithelial cells. Infect. Immun. 67, 4673–4678. Foster, T.J., 2005. Immune evasion by staphylococci. Nat. Rev. Microbiol. 3, 948–958. Fowler, T., Wann, E.R., Joh, D., Johansson, S., Foster, T.J., Hook, M., 2000. Cellular invasion by Staphylococcus aureus involves a fibronectin bridge between the bacterial fibronectin-binding MSCRAMMs and host cell beta1 integrins. Eur. J. Cell Biol. 79, 672–679. Fox, L.K., Zadoks, R.N., Gaskins, C.T., 2005. Biofilm production by Staphylococcus aureus associated with intramammary infection. Vet. Microbiol. 107, 295–299.

K.E. Budd et al. / Veterinary Microbiology 189 (2016) 24–31 Gunther, J., Koy, M., Berthold, A., Schuberth, H.J., Seyfert, H.M., 2016. Comparison of the pathogen species-specific immune response in udder derived cell types and their models. Vet. Res. 47, 22. Haggar, A., Hussain, M., Lonnies, H., Herrmann, M., Norrby-Teglund, A., Flock, J.I., 2003. Extracellular adherence protein from Staphylococcus aureus enhances internalization into eukaryotic cells. Infect. Immun. 71, 2310–2317. Hirschhausen, N., Schlesier, T., Schmidt, M.A., Gotz, F., Peters, G., Heilmann, C., 2010. A novel staphylococcal internalization mechanism involves the major autolysin Atl and heat shock cognate protein Hsc70 as host cell receptor. Cell. Microbiol. 12, 1746–1764. Keane, O.M., Budd, K.E., Flynn, J., McCoy, F., 2013. Pathogen profile of clinical mastitis in Irish milk-recording herds reveals a complex aetiology. Vet. Rec. 173, 17. Lahouassa, H., Moussay, E., Rainard, P., Riollet, C., 2007. Differential cytokine and chemokine responses of bovine mammary epithelial cells to Staphylococcus aureus and Escherichia coli. Cytokine 38, 12–21. Lim, Y., Jana, M., Luong, T.T., Lee, C.Y., 2004. Control of glucose- and NaCl-induced biofilm formation by rbf in Staphylococcus aureus. J. Bacteriol. 186, 722–729. McCarthy, A.J., Lindsay, J.A., 2010. Genetic variation in Staphylococcus aureus surface and immune evasion genes is lineage associated: implications for vaccine design and host-pathogen interactions. BMC Microbiol. 10, 173. Melchior, M.B., Fink-Gremmels, J., Gaastra, W., 2006. Comparative assessment of the antimicrobial susceptibility of Staphylococcus aureus isolates from bovine mastitis in biofilm versus planktonic culture. J. Vet. Med. B Infect. Dis. Vet. Public Health 53, 326–332. Merritt, J.H., Kadouri, D.E., O’Toole, G.A., 2005. Growing and analyzing static biofilms. Curr. Protoc. Microbiol. 1 Chapter 1, Unit 1B. Nguyen, M.T., Kraft, B., Yu, W., Demircioglu, D.D., Hertlein, T., Burian, M., Schmaler, M., Boller, K., Bekeredjian-Ding, I., Ohlsen, K., Schittek, B., Gotz, F., 2015. The nuSaalpha specific lipoprotein like cluster (lpl) of S. aureus USA300 contributes to immune stimulation and invasion in human cells. PLoS Pathog. 11, e1004984. O’Neill, E., Pozzi, C., Houston, P., Smyth, D., Humphreys, H., Robinson, D.A., O'Gara, J. P., 2007. Association between methicillin susceptibility and biofilm regulation in Staphylococcus aureus isolates from device-related infections. J. Clin. Microbiol. 45, 1379–1388. Peton, V., Le Loir, Y., 2014. Staphylococcus aureus in veterinary medicine. Infect. Genet. Evol. 21, 602–615. Peton, V., Bouchard, D.S., Almeida, S., Rault, L., Falentin, H., Jardin, J., Jan, G., Hernandez, D., Francois, P., Schrenzel, J., Azevedo, V., Miyoshi, A., Berkova, N., Even, S., Le Loir, Y., 2014. Fine-tuned characterization of Staphylococcus aureus Newbould 305, a strain associated with mild and chronic mastitis in bovines. Vet. Res. 45, 106.

31

Rachid, S., Ohlsen, K., Wallner, U., Hacker, J., Hecker, M., Ziebuhr, W., 2000. Alternative transcription factor sigma(B) is involved in regulation of biofilm expression in a Staphylococcus aureus mucosal isolate. J. Bacteriol. 182, 6824– 6826. Schukken, Y.H., Gunther, J., Fitzpatrick, J., Fontaine, M.C., Goetze, L., Holst, O., Leigh, J., Petzl, W., Schuberth, H.J., Sipka, A., Smith, D.G., Quesnell, R., Watts, J., Yancey, R., Zerbe, H., Gurjar, A., Zadoks, R.N., Seyfert, H.M., 2011. Host-response patterns of intramammary infections in dairy cows. Vet. Immunol. Immunopathol. 144, 270–289. Smith, E.M., Green, L.E., Medley, G.F., Bird, H.E., Fox, L.K., Schukken, Y.H., Kruze, J.V., Bradley, A.J., Zadoks, R.N., Dowson, C.G., 2005. Multilocus sequence typing of intercontinental bovine Staphylococcus aureus isolates. J. Clin. Microbiol. 43, 4737–4743. Stutz, K., Stephan, R., Tasara, T., 2011. SpA, ClfA, and FnbA genetic variations lead to Staphaurex test-negative phenotypes in bovine mastitis Staphylococcus aureus isolates. J. Clin. Microbiol. 49, 638–646. Thompson-Crispi, K., Atalla, H., Miglior, F., Mallard, B.A., 2014. Bovine mastitis: frontiers in immunogenetics. Front. Immunol. 5, 493. Trouillet, S., Rasigade, J.P., Lhoste, Y., Ferry, T., Vandenesch, F., Etienne, J., Laurent, F., 2011. A novel flow cytometry-based assay for the quantification of Staphylococcus aureus adhesion to and invasion of eukaryotic cells. J. Microbiol. Methods 86, 145–149. Vrieling, M., Koymans, K.J., Heesterbeek, D.A., Aerts, P.C., Rutten, V.P., de Haas, C.J., van Kessel, K.P., Koets, A.P., Nijland, R., van Strijp, J.A., 2015. Bovine Staphylococcus aureus secretes the leukocidin LukMF0 to kill migrating neutrophils through CCR1. MBio 6 00335–00315. Wellnitz, O., Bruckmaier, R.M., 2012. The innate immune response of the bovine mammary gland to bacterial infection. Vet. J. 192, 148–152. Wickstrom, E., Persson-Waller, K., Lindmark-Mansson, H., Ostensson, K., Sternesjo, A., 2009. Relationship between somatic cell count, polymorphonuclear leucocyte count and quality parameters in bovine bulk tank milk. J. Dairy Res. 76, 195–201. Zapotoczna, M., Jevnikar, Z., Miajlovic, H., Kos, J., Foster, T.J., 2013. Iron-regulated surface determinant B (IsdB) promotes Staphylococcus aureus adherence to and internalization by non-phagocytic human cells. Cell. Microbiol. 15, 1026–1041. Zbinden, C., Stephan, R., Johler, S., Borel, N., Bunter, J., Bruckmaier, R.M., Wellnitz, O., 2014. The inflammatory response of primary bovine mammary epithelial cells to Staphylococcus aureus strains is linked to the bacterial phenotype. PLoS One 9, e87374.