Introduction

Coagulase-negative staphylococci (CNS) are a group of bacteria that are increasingly implicated as a cause of hospital-acquired opportunistic and health-care related infections worldwide. Staphylococcus haemolyticus strains are the second of the most frequently encountered CNS that have been recognized as an emerging and important human pathogen causing serious infections such as: endocarditis, urinary tract infections, septicemia, peritonitis, wound, bone and joint infections. Patients with the illnesses are usually immunocompromised, with indwelling or implanted foreign bodies (Piette and Verschraegen 2009).

An important feature of S. haemolyticus strains is resistance to β-lactams and other antibiotics agents (Götz 2006). The treatment of methicillin-resistant staphylococci is generally based on glycopeptides, especially vancomycin. When combination therapy is required, the aminoglycosides are used because of their killing potential and the postantibiotic effect. Significantly, the aminoglycosides have an ability to produce synergistic bactericidal activity in combination with antimicrobial agents inhibiting cell wall biosynthesis, including vancomycin (Vakulenko and Mobashery 2003). In staphylococci resistant to aminoglycosides, it is commonly due to drug inactivation by cellular aminoglycosides-modifying enzymes (You et al. 2000). The bifunctional enzyme AAC(6′)/APH(2″), encoded by aac(6′)/aph(2″) mediates resistance to gentamicin, neomycin, kanamycin, tobramycin and amikacin. The APH(3′)-III enzyme encoded by aph(3′)-IIIa, essentially modified kanamycin and amikacin, whereas the ANT(4′)-I enzyme encoded by ant(4′)-Ia inactivates tobramycin, kanamycin, neomycin and amikacin (Schmitz et al. 1999).

Although S. haemolyticus strains are among the most common CNS species causing hospital-acquired opportunistic infections, little is known about their virulence-associated properties. Takeuchi et al. (2005) analysed the whole genome sequence of human pathogenic strain and reported 57 open reading frames (orfs) associated with virulence. They identified numerous genes encoded putative enzymes and toxins. At least three orfs showed homology to staphylococcal α-hemolysins, Bacillus cereus and streptococcal hemolysins. The S. haemolyticus genome carried genes which encode enzymes for synthesis of the poly-gamma-glutamate capsule that protects against cationic microbiological peptides (Takueschi et al. 2005). The exact role of these cellular and extracellular products in the pathogenesis of S. haemolyticus strains is still unclear.

The putative mechanism of S. haemolyticus pathogenesis remains poorly documented. Therefore in the study, we analysed possible virulence factors of the methicillin-resistant S. haemolyticus clinical strains and assessed the interaction of the bacteria with epithelium. We investigated adherence, invasion, cytotoxic and apoptotic activity of the isolates to HEp-2 cells. Moreover, we determined susceptibility of the strains to antibiotics with a focus on aminoglycoside resistance.

Materials and methods

Bacterial strains and growth conditions

We evaluated 30 methicillin-resistant S. haemolyticus isolates that were previously identified to the species level and clonal analysed by REP-PCR typing (Krzymińska et al. 2012a). The strains originated from clinical samples: blood (15), wounds (4), respiratory secretions (4), skin (2), urine (2) and medical devices (Table 1). They were collected from hospitalized patients over a 2-year period, as described earlier. These strains and Escherichia coli as the negative control were grown overnight in tryptic soy broth (TSB, Difco) 37 °C. The cultures were centrifuged at 2,000×g for 20 min. Supernatants were sterilized through 0.22 μm-pore size membrane filters Millex-GV (Millipore) and heated (56 °C for 20 min) to destroy the activity of heat-labile toxins. For the analysis of cell-contact cytotoxicity, the pellet of bacterial cells were resuspended in PBS and adjusted to the optical density OD of 0.4, corresponding to 0.8–1.5 × 104 colony- forming units (CFU/1 ml).

Table 1 Source of origin and aminoglycoside resistance genes of Staphylococcus haemolyticus strains
Full size table

Antibiotic susceptibility testing

Susceptibility to the following antibiotical agents: ciprofloxacin, clindamycin, erythromycin, gentamycin, tobramycin, oxacillin, rifampicin, teicoplanin, tetracycline, trimethoprim/sulfamethoxazole, vancomycin, levofloxacin, norfloxacin and linezolid was performed using the Vitek 2 system (bioMérieux, France).

Identification of aminoglycoside resistance genes

Isolation of DNA was performed by using the Genomic DNA Plus kit (A&A Biotechnology, Poland). A multiplex PCR was applied for simultaneous amplification of aac(6′)/aph(2″), aph(3′)-IIIa, ant(4′)-Ia gene. The PCR amplification of the mecA genes was performed as described previously by Ardic et al. (2006). The amplification products were electrophoresed in 1.5 % agarose gel, stained with EB, visualized on a UV light transilluminator, and documented with V.99 Bio-Print system (Vilber Lourmat, Torcy, France).

Epithelial cell line

Chinese hamster ovary cells (CHO) and human epidermoid carcinoma cells from the larynx (HEp-2) were cultured in Eagle minimal essential medium (GM, Sigma) with 5 % fetal calf serum (FCS, Sigma) containing 2 mM glutamine, penicillin (50 Iu/ml), streptomycin (100 μg/ml) and 1 mg/ml of nystatin (Krzymińska et al. 2010).

Infection of HEp-2 cells

The epithelial cells were seeded in culture 96-well plates and were grown in GM, until 80 % of confluency was obtained. Next, the medium was replaced with Eagle medium without FCS and antibiotics for 24 h. During infection, the monolayer was incubated with bacterial suspension at a multiplicity of infection (MOI) of 10 bacteria per HEp-2 cell (the cells in number 1 × 105 incubated with approximately of 1 × 106 of bacteria) for 2 h at 37 °C to allow for adhesion and invasion of bacteria to epithelial cells.

Adhesion and invasion assay

Adhesion and invasion of S. haemolyticus to human epithelial cells were quantified using gentamicin, lysostaphin protection assay with modifications (Agerer et al. 2005; Rocha-de-Souza et al. 2008). Monolayer of HEp-2 cells in two sets of triplicate wells was incubated with the bacteria for 2 h. During the incubation the relative growth rate was approximately equal for each strain. To enumerate total associated (adhered and intracellular) bacteria, one set of triplicate wells of infected cells was washed three times with PBS and lysed in 100 μl 0.1 % Triton X-100. For invasion assay, a second set of triplicate wells with infected cells was washed with PBS and incubated with Eagle medium containing 300 μg/ml of gentamicin and 20 μg/ml lysostaphin for 2 h. After washing three times with PBS, the infected cells were lysed with 0.1 % Triton X-100 in PBS. The lysates were serially diluted and plated on TSA and enumerated by colony-forming units (CFU) counting. The number of attached bacteria was determined by subtracting the intracellular bacteria following invasion from the total number of adhered and invaded once. Bacterial adhesion and invasion ability were presented as respectively the adhesion (AdI) and invasion (InI) indexes. The adhesion index was expressed as the mean total number of CFU associated bacteria per well (1 × 105 HEp-2 cells). The invasion index (InI) presented as the number of internalised bacteria per well after gentamicin/lysostaphin treatment. To determine whether the cytoskeleton of epithelial cells is involved in bacterial uptake, the cells were treated with cytochalasin D (to inhibit actin polymerization). In parallel experiments, cytochalasin D (1 μg/ml, Sigma), was added to the cells 30 min prior to infection. Addition of cytochalasin D during the experiment had no effect on the bacterial and epithelial cell viability. The monolayer was infected separately with an invasive strain of Yersinia enterocolitica O:3/4 (pYV+) and non-pathogenic E. coli K-12 C600.

Cell-contact and extracellular toxic activities

Cytotoxic activity to epithelial cells was measured in a bacterial cell suspension and extracellular culture filtrates. In order to determine cell-contact cytotoxicity, epithelial cells were grown in six-well plates containing 1 ml of GM per well. The bacterial cell suspension in GM at a MOI of 10:1 was added to each well and separately to transwell cell culture chamber inserts (Nunc). The membrane at the bottom of the insert precludes direct interaction between bacteria and epithelial cells. However, it allows for the exchange of extracellular products. The monolayer of HEp-2 cells was incubated with the bacterial cell suspension for 5 h and cell-free culture supernatant for 5 and 24 h at 37 °C. The cytotoxic activity was assessed in Neutral Red (NR) retention assay (Krzymińska et al. 2010). Cytotonic activity of cell associated and extracellular factors was observed on CHO cells.

Assessment of apoptosis

Apoptosis of infected epithelial cells was assessed by three different methods: trypan blue exclusion assay, double staining with acridine orange (AO, 100 μg/ml)/ethidium bromide (EB, 100 μg/ml) and an analysis of nuclear DNA fragmentation. Viability of the cells was tested in the trypan blue exclusion assay (Krzymińska et al. 2010). Morphological evidence of apoptosis and necrosis was obtained by means of AO and EB staining (Krzymińska et al. 2012a, b). The cells were examinated with a Zeiss Axiovert 405 M inverted microscope equipped with epifluorescence. Confocal images of green fluorescence were collected using a 488 excitation and 514–540 band-pass filter. Images of red fluorescence were collected using a 568 nm excitation and 590 long pass filter. Images were acquired using a Zeiss AxioCam HRc camera operating under AxioVision 3,1 software, and next processed using Adobe Photoshop CS5. The cells were recorded in three different groups: viable, apoptotic and necrotic cells. Viable (green), apoptotic (green with red nuclei) and necrotic cells (red) were quantified by counting a minimum of 100 cells in three independent experiments. The results were presented as the percentage of apoptotic cells [Apoptotic Index (AI)] and the percentage of necrotic cells (Necrotic Index). A biochemical hallmark of apoptosis is fragmentation of DNA that was extracted from the infected cells (Krzymińska et al. 2012b).

Transmission electron microscopy (TEM)

Electron microscopy was used to evaluate morphological changes of HEp-2 cells incubated with S. haemolyticus strains. The cells, after 3 h-infection, were washed three times with PBS and harvested by trypsinization. Next, they were centrifuged at 1,500×g for 15 min, fixed with 2.5 % glutaraldehyde in 0.1 M phosphate buffer for 1 h, washed with phosphate buffer, postfixed with 1 % OsO4, gradually dehydrated in series of acetone, and embedded in Epon (Choi et al. 2005). The samples were sliced into 70-nm sections, stained with uranyl acetate and examined with a transmission electron microscope (JEM 1200 EXII) TEM at accelerating voltage of 80 kV.

Determination of exoenzymes activity

We analysed the activity of extracellular enzymes that may act as putative virulence factors: lipase, lecithinase (phospholipase) and serine protease. Lipase activity was detected on TSB agar (TSA) enriched with 0.01 % CaCl2 × 2H2O and 1 % Tween 80 (Difco). Lipolytic activity was indicated by the appearance of a precipitation zone resulting from the deposition of crystals formed with calcium salt and fatty acids generated by the enzyme (Kumar et al. 2012). Lecithinase activity was examined on BHI agar supplemented with 10 % (vol/vol) egg-yolk (Karasawa et al. 2003). Inoculated plates were incubated for 48 h at 35 °C and 24 h at room temperature. Lecithinase activity was characterized by a transparent zone around colonies. Production of serine protease was determined on TSA supplemented with 5 % (w/v) casein (Lucas and Manna 2013). The activity was observed as a transparent zone surrounding the bacteria colonies.

Statistical analysis

The data were presented as mean ± standard deviation from two independent experiments performed in duplicate. A one-way analysis of variance ANOVA with Tukey’s post hoc test at the significance level P < 0.05 was performed. The linear regression analysis was used to examine pairwise correlation between the AI, cytotoxicity, the Pearson correlation coefficient was determined. P values of <0.05 were considered statistically significant. The statistical analysis was performed using Statistica PL software (StatSoft Poland Inc., USA).

Results

Antibiotic resistance of S. haemolyticus strains

All S. haemolyticus strains were resistant to oxacillin. The mecA gene was present in all the strains. Resistance to vancomycin, teicoplanin and linezolid was not detected. Analysis of the antibiotic resistance of the strains showed that all of them were multiresistant i.e., expressing resistance to more than three different classes of antibiotics, in addition to β-lactams. Erythromycin resistance was found for 28 (93 %) strains. We observed resistance to gentamycin for 40 % of strains, tobramycin (33 %), trimethoprim/sulfamethoxazole (56 %), clindamycin (50 %) and tetracycline (36 %). Only two strains (7 %) exhibited resistance to rifampicin. Resistance to ciprofloxacin, norfloxacin and levofloxacin was revealed respectively by 53, 36 and 30 % of the strains.

The presence of aminoglycoside resistance genes

The PCR analysis revealed that aminoglycoside resistance was due to the existence of gene aph(3′)-IIIa for 43 % of the strains, aac(6′)/aph(2″) (33 %), and ant(4′)-Ia (20 %). Ten isolates (33 %) contained two AME genes, aac(6′)/aph(2″) and aph(3′)-IIIa, simultaneously (Table 1). At least one AME was encountered in 19 (63 %) of S. haemolyticus strains, whereas 11 strains did not harbour any of the AME. It is noteworthy that among the 12 gentamicin-resistant strains, two strains did not harbour aac(6′)/aph(2″). All gentamicin-susceptible S. haemolyticus isolates were negative by PCR for the bifunctional gene.

In Table 2 we present summarized results of assessment of S. haemolyticus apoptotic, cytotoxic activities, adhesion, invasion to HEp-2 cells, and enzyme activity that might contribute to the virulence.

Table 2 Apoptotic index, cytotoxic, adhesion, invasion activities and production of lipase, lecithinase and proteinase by S. haemolyticus strains
Full size table

S. haemolyticus adherence to epithelial cells

All of the S. haemolyticus strains adhered to human epithelial cells with the range from 2 × 105 to 1.53 × 107 CFU per 1 × 105 cells. The highest adhesion index, ranging from 1.31 to 1.53 × 107 was revealed for 6 (20 %) strains. The lowest adhesion ability, from 0.2 × 106 to 1.1 × 106 CFU was observed for 7 (23.3 %) isolates. The negative control of nonpathogenic E. coli K12C600 showed the index of 1.2 × 102 CFU, whereas that of Y. enterocolitica O:8/1B, positive control elevated to 25.8 × 105 CFU.

S. haemolyticus invasion of epithelial cells

Nineteen strains (63.3 %) exhibited invasion of epithelial cells. The percentage of associated bacteria that were internalized ranged from 0.01 to 17.9 %. Three (10 %) strains showed the highest invasion activity with the invasion index from 17.9 to 12.6 %. Six (20 %) isolates exhibited the lowest invasive ability (0.01–0.07 %), higher than that of the nonpathogenic negative control. Eleven (37 %) strains did not reveal invasive ability. The percentage of invading bacteria ranged from 0.1 to 10.4 %, which was higher than the negative control, was observed for 10 (64 %) isolates. The invasion index for Y. enterocolitica O:8/1B reached 57.3 %, whereas for E. coli K12C600 strains that invaded HEp-2 cells, it was 0.003 % of total adhered cells. Treatment of epithelial cells with cytochalasin D resulted in a reduction of the bacteria uptake in the range from 84 % (±1.7 %) to 91 % (±1.9 %) after 24 h. These data suggested that actin polymerization plays a major role in bacterial internalization. Treatment of epithelial cells with cytochalasin D had no effect on S. haemolyticus adhesion to epithelial cells.

Cell-contact cytotoxicity of S. haemolyticus strains

To determine whether cell contact is required for S. haemolyticus-induced cytotoxicity, infected epithelial cells were observed using inverted phase-contrast microscopy. After 5-h incubation, the cells were detached from the plate surface (results not shown). Morphological changes were confirmed by the NR test (Table 2). The cell contact-cytotoxicity was ranging from 9.6 to 66.7 %. The highest activity in the range from 53 to 66.7 % of destructed cells was revealed by 10 (30 %) strains. The lowest cytotoxicity below 10 % was observed for 1 (3.3 %) strain. Low percentage of cytotoxicity from 0.4 to 1.1 could be noticed when S. haemolyticus cells were not allowed to contact with epithelial cells in the culture inserts. The results suggested that extracellular factors may contribute to the cytotoxicity. Bacterial culture supernatants of 19 (63.3 %) strains showed a low effect (7.3 ± 2.1 %) on viability of HEp-2 cells at 5-h incubation. No cell lysis was observed at 5-h incubation with culture medium (GM) and non-pathogenic E. coli K-12 C600 strain.

Extracellular cytotonic and cytotoxic activity

None of S. haemolyticus isolates revealed cytotonic activity to CHO cell line, whereas all strains revealed activity of extracellular factors which destroy HEp-2 cells. The percent of cytotoxicity ranged from 13.8 to 81.5 % at 24 h. The highest activity in the range from 72.8 to 81.5 % was observed for 8 (26.7 %) isolates. Seven (23.3 %) strains exhibited the lowest activity in the range from 11.2 to 25.7 %. The negative control E. coli K-12 C600 strain was not cytotoxic to epithelial cells. Preheating (56º C for 20 min) of cell-free supernatants of 23 (77 %) strains caused a reduction of cytotoxic activity to HEp-2 cells in the range from 7.1 % (±1.7 %) to 19.6 % (±2.9 %) after 24 h.

Apoptosis of epithelial cells induced by S. haemolyticus strains

Infected epithelial cells with S. haemolyticus isolates revealed signs of apoptosis in the EB/AO staining. AO penetrates into viable cells and stain them green (Fig. 1a). EB is taken up by the cells when cytoplasmic membrane integrity is lost and colour the nuclei red (Fig. 1b). Viable (green), apoptotic (green with red nuclei) and necrotic cells (red) were quantified by counting a minimum of 100 cells in three independent experiments. Immunofluorescence photography also showed bacteria adhered and invaded to epithelial cells.

Fig. 1
figure 1

Apoptosis of HEp-2 cells due to S. haemolyticus infection. The cells were stained with AO/EB and observed under a laser confocal microscope. a—control, noninfected cells b—epithelial cells infected with S. haemolyticus MPU Sh4 at 48 h after infection; a—live cells, b—apoptotic cells, c—necrotic cells, d—intracellular bacteria

Full size image

All S. haemolyticus strains induced epithelial cell death through apoptosis. The AI ranged from 1.2 to 59.8 % at 24 h after infection. The highest AI (46.2–59.8 %) at 24 h was observed in cells incubated with 6 (20 %) strains. The lowest AI, between 1.2 and 9.1 %, was expressed by 7 (23.3 %) isolates. The percentage of apoptotic cells increased in the range from 2.2 to 89.8 % at 48 h post infection. The highest AI ranging from 76.2 to 89.8 % was observed in HEp-2 cells infected with 5 (16.7 %) of the S. haemolyticus strains. The lowest AI ranging from 2.2 to 21.3 % was revealed by cells infected with 7 (23.3 %) of the strains. The mean AI of the negative control was 4.1 ± 1 %, whereas for the positive control it reached 94.6 ± 5.1 %. Some S. haemolyticus strains also exhibited necrotic activity. The highest necrotic index was observed for six strains (11.3 %) at 24 h and 8 isolates (15.1 %) at 48 h. Their indexes ranged from 39 to 41 % and from 45 to 69 %, respectively. Treatment of HEp-2 cells with cytochalasin D prior to infection, resulted in reduction of the apoptotic indexes to the range from 8.1 to 3.7 % for 19 strains (63 %) that exhibited invasion ability. Apoptosis of epithelial cells was also confirmed by the analysis of DNA fragmentation in cells infected with S. haemolyticus strains (Fig. 2). In the present study 2 (6.7 %) strains induced the intranucleosomal pattern with the size of about 180–200 bp in DNA extracted from cells at 24 h after infection. The number of strains that caused DNA fragmentation in HEp-2 cells increased to 17 (56.7 %) after 48 h of infection. Nuclear DNA fragmentation was observed when AI exceeded 51 %. The positive control of Y. enterocolitica O:8/1B induced DNA fragmentation in epithelial cells, whereas non-pathogenic E. coli K-12 C200, 28 (93.3 %) and 13 (43 %) of S. haemolyticus strains did not induce nuclear DNA fragmentation, respectively at 24 and 48 h post infection.

Fig. 2
figure 2

Intranucleosomal degradation of DNA from HEp-2 cells infected with different strains. DNA isolated from cells infected with: lane 1—E. coli K12C600 (negative control); lane 2—S. haemolyticus MPU Sh7 at 24 h after infection; lane 3—S. haemolyticus MPU Sh7 at 48 h after infection; lane 4—S. haemolyticus MPU Sh4 at 24 h after infection; M—molecular size marker

Full size image

Electron microscopic examination of infected epithelial cells

Transmission electron microscopy revealed that the negative control had an intact cell and organelle membrane and organelles and large nuclei (Fig. 3a). Infected with S. haemolyticus epithelial cells disclosed signs of apoptosis. The cells had an intact cell and nuclear membrane (Fig. 3b, c). The cells revealed chromatin condensation and decreasing of nuclear volume and its fragmentation (Fig. 3b), formation of apoptotic bodies inside the cells (Fig. 3b, c). The condensed chromatin in the apoptotic cells is localized to periphery (Fig. 3c) or one side of the nuclear fragments (Fig. 3b). Electron microscopy revealed that S. haemolyticus strains during adhesion, induced the formation of pseudopod structures of the epithelial cell wall, at the site of the bacterial attachment. The intracellular bacteria could be found residing in the membrane bound vesicles inside the cells. The vacuoles contained single bacteria (Fig. 3c).

Fig. 3
figure 3

Transmission electron micrographs of epithelial cells. a—control, noninfected HEp-2 cell; b—chromatin condensation and nuclear condensation and fragmentation in cells infected with S. haemolyticus MPU Sh4 at 24 h; c—invaded bacteria into cellular cytoplasm were detected within endosomal vacuoles (a)

Full size image

Activity of exoenzymes

Among thirty S. haemolyticus strains, eight (27 %) revealed lipolytic activity, 13 (43 %) produced lecithinase and 6 (20 %) were positive for proteinase activity.

Discussion

The mechanisms of pathogenesis and virulence factors of S. haemolyticus strains have not yet been well defined. Therefore, this study was important for clarifying the pathogenicity of S. haemolyticus clinical isolates.

In this study, we observed a high rate of aminoglycoside resistance (63 %) of methicillin-resistant S. haemolyticus strains. The aph(3′)-III was found in 43 % and interpreted as the most frequently carried AME genes. Liakopoulos et al. (2011) suggested that the aph(3′)-III was also the most prevalent AME gene among Staphylococcus aureus and CNS isolated in Greece. In contrast, Ida et al. (2001) revealed that the aph(3′)-III genes were carried only by 8.9 % strains originated from Japan, whereas the most prevalent was the ant(4′)-Ia gene. In our studies, the ant(4′)-Ia gene was present only in six (20 %) strains. The aac(6′)/aph(2″) gene encoding bifunctional enzyme was present in 33 % of tested S. haemolyticus strains. In studies of staphylococci from 19 European hospitals, the aac(6′)/aph(2″)gene was present in 68 % isolates and was the most prevalent (Schmitz et al. 1999). The aac(6′)/aph(2″) genes have been found in S. haemolyticus isolated from human breast milk (Carneiro et al. 2004).

The epithelium is the first line of defence against infectious agents, playing a broad range of protective roles in the innate response to infection. Bacterial adherence to epithelial cells is a crucial step for pathogens to initiate and establish host colonization or to direct impact upon the eukaryotic cells (Hoffmann et al. 2011; Kim et al. 2010). Our results demonstrated that all isolates were able to adhere to HEp-2 cells. For 30 % isolates, the adhesion index was comparable to that of Y. enterocolitica O:8/1B, the enteroinvasive positive control. The highest adhesion ability was observed for 7 (67 %) strains isolated from blood and one isolate from skin and wound infections. Hyvönen et al. (2009) demonstrated that S. haemolyticus isolated had an adhesive ability to bovine mammary epithelial cells.

The ability of pathogenic bacteria to invade host cells and tissues is considered one of the most critical pathogenicity factors in persisting infections. Invasion of host cells allows pathogens to hide, persist within host tissues and to escape immune response (Hoffmann et al. 2011). This study revealed that S. haemolyticus strains were internalized by non-professional phagocytes. We observed that 19 of 30 strains (63 %) were invasive, with the invasion index higher than that of the nonpathogenic control. The highest invasion index, approximately from three to four times lower than that of Y. enterocolitica O:8/1B was observed for isolates originated from aspirate (MPU Sh2) and wounds (MPU Sh9, 4). Six isolates (20 %) revealed the lowest invasive ability, with efficiency higher than that of the E. coli nonpathogenic control. The results of S. haemolyticus invasion of epithelial cells were confirmed in transmission electron microscopy. We observed intracellular localization of the bacteria inside cytoplasmic vacuoles. Internalized bacteria can evade the host immune system and antibiotic treatment. The survival of S. haemolyticus strains in non-phagocytic cells, such as epithelial cells could have implication in bloodstream infections provoked by the isolates. It should be noted that 50 % of S. haemolyticus strains were isolated from blood samples (Krzymińska et al. 2012a). Szabados et al. (2008) suggested that Staphylococcus saprophyticus is internalized by the human urinary bladder carcinoma cell line. Foster et al. (2014) have identified a broad range of proteins that are covalently attached to peptidoglycan, which are known as cell wall-anchored proteins. These proteins revealed numerous functions, including adhesion to and invasion of host cells, biofilm formation and evasion of the host immune response. Hirschhausen et al. (2010) reported a novel mechanism of CNS internalization of the host cells, which is mediated by autolysins/adhesins from Staphylococcus epidermidis (AltE). The proteins have enzymatic and adhesive functions. In our results, the percentage of invading bacteria decreased after treatment of cytochalasin D prior to infection. The results suggested that invasion of S. haemolyticus strains to epithelial cells could be similar to phagocytosis that involved actin polymerization and subsequently the bacteria internalization into a vacuole. Sinha and Fraunholz (2010) reported that S. aureus revealed intracellular persistence in host cells. Whereas the bacteria were regarded as extracellularly located strains, some reports suggested their intracellular localization in host cells. The persistence of S. aureus in host cells is associated with their localization in vacuolar endosomal compartments. Internalization of S. aureus was inhibited by addition of cytochalasin D (Szabados et al. 2011).

We demonstrated that one of the virulence pathways of S. haemolyticus strains provoked contact-dependent cytotoxicity. In the study, all the strains displayed this activity, whereas little or no cytotoxicity was detected at the same time in culture supernatant of the strains. The results indicated that S. haemolyticus strains promote epithelial cell lysis by a virulence factor associated with a bacterial cell. We observed that 27 % of the strains revealed lipolytic activity, 43 % produced lecithinase and 20 % were positive for proteinase activity. We observed that strains that produced lipase, lecithinase and proteinase activities (MPU Sh4) or lipase and proteinase (MPU Sh9, 11, 13, 5, 2) exhibited the highest cell-contact cytotoxicity and invasion index. Some pathogenic strains may produce extracellular enzymes that could be responsible for host tissue degradation and development of an infection caused by the bacteria. Proteases can degrade host complement factors, mucins, and disrupt tight junctions between epithelial cells, leading to dissemination of the bacteria (Otto 2004). Proteases have been reported to be able to degrade fibrinogen, complement and other proteins, suggesting a role in the escape from innate defence system. Some CNS produced extracellular metalloprotease with elastase activity. The elastase that degrades human sIgA, IgM, serum albumin, fibronectin and fibrinogen is assumed to be a virulence factor (Otto 2004; Von Eiff et al. 2002). Lipases and phospholipases can target lipids in surfactants and host cell membranes. The enzymes may contribute to virulence by enabling the bacteria to persist in fatty secretions of human skin.

We observed that all S. haemolyticus strains exhibited cytotoxic activity of cell-free culture supernatant at 24 h after infection which suggested production of extracellular toxins. The highest activity in the range from 72.8 to 81.5 % was observed for 26.7 % isolates. Culture filtrates of the strains incubated at 56 °C for 20 min revealed a decrease in cytotoxic activity to epithelial cells that suggested that they produce heat-labile toxins. The results suggested that the strains displayed virulence factors involved in the infection process, showing the ability to damage host tissues as well as to evade the host defence system. Disruption of the epithelial barrier integrity may favour the bacterial tissue invasion and entry into the bloodstream. Previously, we noted that all S. haemolyticus strains were cytotoxic to macrophages (Krzymińska et al. 2012a). The important virulence factors that exert cytotoxic activity of CNS strains are extracellular enzymes and toxins: metalloprotease with elastase activity, cysteine and serine proteases, lipase, FAME and δ-toxins (Piette and Verschraegen 2009).

Our study revealed the fate of epithelial cells after bacterial adhesion and internalization. We demonstrated that one of the consequences of cell adhesion, invasion, and cytotoxic activity by S. haemolyticus strains is injury to the cells and cell death by apoptosis. We observed morphological changes of infected epithelial cells, including condensation of nucleus and nuclear chromatin, formation of apoptotic bodies. The highest apoptotic activity was revealed by strains isolated from wounds (MPU Sh4, 9) and urine (MPU Sh3). We observed that 56.7 % of strains caused fragmentation of nucleosomal DNA of infected cells to multimers which are a biochemical standard of apoptosis. The analysis of morphological changes and internucleosomal cleavage of the host cell DNA indicated that six strains isolated from blood, three from wounds and secretions, three from urine and medical devices and one strain originated from skin caused apoptosis of epithelial cells. All the strains which induced apoptosis of more than 58.4 % of epithelial cells were multiresistant to more than three different classes of antibiotics. Moreover, five strains revealed aminoglycoside resistance due to existence of genes aac/aph (MPU Sh2, 11, 25) and aph(3′)-IIIa (MPU SH10, 19). Seven strains did not harbor AME. The Pearson linear test revealed positive correlations between the AI of infected epithelial cells at 24 h and percentage of cell-cytotoxicity (r = 0.5, P < 0.01), extracellular cytotoxicity (r = 0.6, P < 0.01) and the invasion index (r = 0.2, P < 0.01). Current evidence suggested that apoptotic epithelial cells activate molecular signalling pathways for recognition by professional phagocytes and next removal of dead cells (Elliot and Ravichandran 2010).

The exact contribution of the S. haemolyticus virulence factors to pathogenesis is unclear. The present study showed that the importance of the strains as an emerging cause of nosocomial infections is connected with multiple strategies including antibiotics resistance, adhesion and invasion of human epithelial cells, production of extracellular enzymes and toxins that interact with the host cells. Moreover, the strains revealed evasion mechanisms and a survival strategy, residing in membrane-bound compartments and blocking essential non immune-mediated functions during infection.