Staphylococcus aureus is a commensal organism and a frequent cause of skin and soft tissue infections, which can progress to serious invasive disease. This bacterium uses its fibronectin binding proteins (FnBPs) to invade host cells and it has been hypothesised that this provides a protected niche from host antimicrobial defences, allows access to deeper tissues and provides a reservoir for persistent or recurring infections. FnBPs contain multiple tandem fibronectin-binding repeats (FnBRs) which bind fibronectin with varying affinity but it is unclear what selects for this configuration. Since both colonisation and skin infection are dependent upon the interaction of S. aureus with keratinocytes we hypothesised that this might select for FnBP function and thus composition of the FnBR region. Initial experiments revealed that S. aureus attachment to keratinocytes is rapid but does not require FnBRs. By contrast, invasion of keratinocytes was dependent upon the FnBR region and occurred via similar cellular processes to those described for endothelial cells. Despite this, keratinocyte invasion was relatively inefficient and appeared to include a lag phase, most likely due to very weak expression of α5β1 integrins. Molecular dissection of the role of the FnBR region revealed that efficient invasion of keratinocytes was dependent on the presence of at least three high-affinity (but not low-affinity) FnBRs. Over-expression of a single high-affinity or three low-affinity repeats promoted invasion but not to the same levels as S. aureus expressing an FnBPA variant containing three high-affinity repeats. In summary, invasion of keratinocytes by S. aureus requires multiple high-affinity FnBRs within FnBPA, and given the importance of the interaction between these cell types and S. aureus for both colonisation and infection, may have provided the selective pressure for the multiple binding repeats within FnBPA.
Citation: Edwards AM, Potter U, Meenan NAG, Potts JR, Massey RC (2011) Staphylococcus aureus Keratinocyte Invasion Is Dependent upon Multiple High-Affinity Fibronectin-Binding Repeats within FnBPA. PLoS ONE 6(4): e18899. doi:10.1371/journal.pone.0018899
Editor: Malcolm James Horsburgh, University of Liverpool, United Kingdom
Received: December 20, 2010; Accepted: March 11, 2011; Published: April 22, 2011
Copyright: © 2011 Edwards et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: JRP acknowledges the British Heart Foundation (http://www.bhf.org.uk/) for a Senior Basic Science Research Fellowship. This work was funded by the Wellcome Trust (http://www.wellcome.ac.uk/), grant number WT 079588. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Staphylococcus aureus is a bacterium responsible for a wide range of superficial and invasive infections ranging in severity from mild to fatal . In addition to causing severe morbidity and mortality in the healthcare environment, S. aureus is a growing problem in the community, causing serious infections in otherwise healthy people , . Treatment of S. aureus infections is often complicated by the high prevalence of antibiotic resistant strains , . Despite the ability of this organism to cause serious illness, S. aureus is primarily a commensal organism, residing within the nares and on the skin of 20–60% of the population either permanently or transiently . Colonisation of the skin can lead to a number of persistent or recurring infections including, folliculitis, scalded skin syndrome, impetigo, colonisation of indwelling medical devices and wound infections , , .
Although originally considered an extracellular pathogen, there is both in vitro and in vivo evidence that S. aureus invades host cells. Although the role of invasion in colonisation and infection is unclear, it is hypothesised to facilitate evasion of immune surveillance, traversal of cellular barriers, evasion of antimicrobial therapy and to enable persistent infection –. Indeed, there is evidence that S. aureus is able to dramatically alter its phenotype (to the small colony variant phenotype) to enhance survival within host cells, which is associated with persistent infections , .
The primary mechanism by which S. aureus enters host cells is well characterised; staphylococcal fibronectin binding proteins (FnBPs) interact with cell surface α5β1 integrins via a fibronectin bridge –. It appears that FnBPA alone is sufficient for invasion since heterologous expression on the surface of otherwise non-invasive Lactococcus lactis or Staphylococcus carnosus confers the ability to invade host cells . The indirect interaction of FnBPA with α5β1 integrins leads to cell signalling events, actin rearrangement and internalization of the bacterium via a mechanism that is entirely dependent on host-cell processes , , .
FnBPs are multifunctional proteins, comprised of distinct regions with variable binding activity. The N-terminal domain binds both fibrinogen and elastin and is implicated in biofilm formation , . This region is followed by 11 (FnBPA) or 10 (FnBPB) non-identical fibronectin-binding repeats (FnBRs), with either high or low-affinity for fibronectin . These multiple repeats enable a single FnBPA molecule to bind multiple fibronectin molecules –. It has been hypothesised that this facilitates bacterial interactions with multiple integrins, triggering cell signalling processes, actin rearrangement and bacterial internalisation –. We have previously shown that this region is essential for triggering bacterial invasion of endothelial cells , . In addition to its role in adhesion, invasion and biofilm formation, the high prevalence of fnb genes amongst S. aureus strains suggest that FnBPs might be important for colonisation; analysis of a panel of 163 clinical isolates revealed that 22% encoded just fnbA, 1% just fnbB and 77% encoded both genes .
We recently investigated how the composition of the FnBR region of FnBPA affected the invasion of endothelial cells and virulence in a murine bacteremia model . This study demonstrated that a single high-affinity FnBR was sufficient to trigger invasion, although this was less efficient than FnBPA variants containing multiple FnBRs . Multiple FnBRs were required for virulence, enhancing bacterial dissemination into the kidneys, as well as leading to significant weight loss and death . Although it is important that we understand how this protein contributes to invasive infection, this is a relatively rare event. It is more likely that the frequent interactions that occur between S. aureus and keratinocytes, involved in colonisation and infection of both nasal and skin surfaces , , , are responsible for selection of FnBPA function. We therefore investigated the role of the FnBR-region in the adhesion to, and invasion of, keratinocytes.
Bacterial strains and growth conditions
A detailed list of the strains used in this study is presented in Table 1. S. aureus 8325.4 strains were cultured in Tryptic Soy Broth (TSB, Oxoid) at 37°C in air with shaking for 16 h. S. aureus CFU were quantified on Tryptic Soy Agar (TSA, Oxoid) plates incubated overnight at 37°C in air. L. lactis strains were cultured in M17 broth (supplemented with 0.5% w/v glucose) for 16 h at 30°C in air (with the indicated concentration of nisin). As required, bacteria were incubated in the presence of the following antibiotics: Chloramphenicol 10 µg ml−1 (S. aureus) or Erythomycin 5 µg ml−1 (L. lactis) or 250 µg ml−1 (E. coli).
Table 1. Strains and plasmids used in this study.doi:10.1371/journal.pone.0018899.t001
S. aureus 8325.4-derived ΔfnbA/B expressing fnbA variants from a plasmid have been characterised previously  to establish equal surface expression levels between strains expressing each construct. All constructs have identical N-terminal and C-terminal domains, varying only in the composition of the FnBR-region.
Recombinant FnBR expression
A polypeptide corresponding to high-affinity repeats FnBR9,10 (FnBPA residues 763–838; Swiss-Prot entry P14738) was expressed and purified in a manner similar to that previously described for single FnBRs . The GST-tag was removed using 3C protease and FnBR9,10 was further purified by reversed-phase HPLC.
Construction and controlled expression of FnBPA variants in L. lactis
Expression levels of FnBPA variants in S. aureus and L. lactis have been determined previously . With the exception of the pMR6-8 construct, all nisin-inducible FnBPA constructs have been described previously . The pMR6-8 construct was made as described previously using standard molecular biology techniques , . Briefly, the enitre R6-8 fnbA variant was amplified from pFnBPR6-8 using primers containing NcoI and XhoI sites (AAACCATGGAGGAGGTATTATAGTGAAAAACAATCTTAGG) (AAACTCGAGCTAACTTTATCTCTCAGTTCGTTATC) and ligated into similarly digested pMSP7517 plasmid. Ligated constructs were transformed into CaCl2 treated E. coli K12 ER2925. Plasmids were recovered from the E. coli transformants (the DNA sequence was confirmed) and transformed into L. lactis NZ9800. As all constructs contained the A domain (Table 1), expression was confirmed by assessment of fibrinogen binding . Previous work has determined that nisin controlled expression of FnBPA variants in L. lactis using this system produces equal expression levels between strains at identical nisin concentrations .
The spontaneously immortalised keratinocyte cell line HaCat was used because it closely resembles primary keratinocyte cells and has been used previously in studies of S. aureus-keratinocyte interactions –. All cells were cultured in Dulbecco's modified Eagle's medium (Invitrogen), with calcium adjusted to 2.8 mM and supplemented with FBS (10%) and l-glutamine (2 mM) at 37°C and 5% CO2, conditions that support differentiation, keratinization and tight-junction formation .
Cells were cultured in T75 flasks to approximately 95% confluency, liberated with trypsin-EDTA, resuspended in culture medium and added to 24-well plates containing thermanox glass coverslips . Plates were incubated for 48 h as described above before the coverslips were removed, dip washed in (phosphate buffered saline)PBS and added to new 24-well plates containing fresh medium and bacteria . For experiments using inhibitors of cell function, these were incubated with the cells 60 min prior to the addition of bacteria and concentrations maintained during the assay; Genistein (200 µM), Wortmannin (20 nM), cytochalasin D (50 µM), PP2 (10 µM), cholchicine (1 µM), cycloheximide (25 µM) and methyl-β-cyclodextrin (2 mM) . EA.Hy926 cells were cultured and prepared as previously described .
Cell attachment and invasion assays
Attachment and invasion assays were performed as described before . Cultured cells were dissociated from plastic flasks using trypsin-EDTA solution (Invitrogen) and approximately 5×105 (in 0.5 ml medium) were seeded into each well of 24-well plates (Nunc) containing 13 mm plastic Thermanox™ cover slips (Fisher) and allowed to attach for 48 h (37°C, 5% CO2). Coverslips were dip-washed once in PBS and placed in the well of a new 24-well plate containing 490 µl of DMEM containing 10% FBS. To each well, 10 µl of washed bacteria were added (approximately 1×107 CFU S. aureus and 5×107 L. lactis) and incubated for 15–90 minutes at 37°C in 5% CO2.
To measure the total number of bacteria associated with the cells (adherent and internalized – referred to in the text and figures as adhesion), coverslips were dip-washed three times in PBS and added to wells containing 500 µl 0.5% Triton X-100. Wells containing coverslips were agitated by pippetting to fully lyse the cells and CFU were enumerated by serial dilution and plating onto TSA agar plates and incubation overnight at 37°C.
For invasion assays, the bacterial suspension was removed and replaced with 500 µl DMEM/10% FBS supplemented with 200 µg ml−1 gentamicin and incubated at 37°C in 5% CO2 for 60 min. Coverslips were washed three times in PBS, lysed and plated for CFU as described for the adhesion assay above. In assays where metabolic inhibitors were used, these were added to cell monolayers for 60 min prior to the experiment and concentrations maintained during incubation with bacteria.
Determination of α5β1 integrin expression by keratinocytes and endothelial cells
HaCat Keratincoytes and EA. Hy926 endothelial cells were cultured in T75 flasks to approximately 95% confluency, as described above. Cells were liberated from the flasks by the use of non-enzymatic cell dissociation solution (4 ml, Sigma) and collected by centrifugation (2,000× g, 10 min). Cells were subsequently resuspended in lysis buffer (1% TX-100, 100 units ml−1 DNase, protease inhibitor cocktail (Sigma) in PBS) and incubated for 20 min at 37°C. Insoluble debris was pelleted by centrifugation and the supernatant recovered. Total protein was determined using the Bradford assay and 20 µl aliquots containing 10 µg protein mixed with Laemmli sample buffer  before boiling for 5 min. Samples were then subjected to SDS-PAGE on duplicate 10% acrylamide gels. Total protein was detected by the use of EZ-blue protein detection reagent (Sigma). Alternatively, proteins were transfered to nitrocellulose membrane using a Bio-Rad semi-dry blotter. The membrane was subsequently blocked in PBS containing 10% skimmed milk powder for 1 hr at room temperature. The protein bands corresponding to α5 or β1 integrin subunits were detected using rabbit polyclonal anti-integrin α5 (H-104) or β1 (M-106, both from Santa Cruz Biotechnology) antibodies and then HRP-conjugated protein A (Sigma). Bound antibody/protein A was detected using enhanced chemiluminescence reagent (GE Healthcare).
Scanning electron microscopy
S. aureus was incubated with keratinocytes or endothelial cells as above for invasion assays for 15–90 minutes. Coverslips were then washed twice with serum-free DMEM before fixation with DMEM containing 2.5% gluteraldehyde and 10 mg ml−1 potassium ferrocynaide for 2 hr at 37°C in 5% CO2. Cells were postfixed in a solution of 1% osmium tetroxide and 1% potassium ferrocyanide for 1 hr in air at room temperature. Coverslips were washed twice in distilled water over 10 mins before staining with 2% uranyl acetate for 1 hr in the dark. Cells were subsequently dehydrated in increasing concentrations of acetone (50–100%) followed by 1:1 acetone: hexamethyldisilazane and finally 100% hexamethyldisilazane, which was allowed to evaporate over 1–2 hrs in air. Samples were then examined using a JEOL JSM6480LV scanning electron microscope.
For adhesion and invasion assays, statistical analyses were performed with Student's t test using the Bonferroni correction for multiple comparisons . Values that were statistically significantly different from controls are indicated by asterisks in the figures. Error bars indicate the mean average ± standard deviation of multiple independent experiments (indicated in the figure legend).
S. aureus 8325.4 adhesion to keratinocytes is similar to endothelial cells but invasion levels are significantly lower
S. aureus attachment to the skin is the precursor to colonisation and infection. Invasion is also likely to play an important role, particularly in maintaining persistent or recurring infection . We employed the well-characterised keratinocyte cell line HaCat as a model to study the role of FnBPA in bacterial adhesion to, and invasion of, the skin and nares. To determine the kinetics of keratinocyte-S. aureus interactions, we measured adhesion to and invasion of HaCat keratinocytes over time, until saturation levels were observed (Fig. 1). Adhesion occurred rapidly, with approximately 3×105 CFU (~3% inoculum) S. aureus attached after 15 minutes (Fig. 1). This number did not change significantly over time (up to 90 mins), suggesting that all available binding sites on the host cells were occupied (Fig. 1). Invasion, by contrast, occurred more slowly. After 15 mins <103 CFU S. aureus had internalised, despite the high number of adherent bacteria and there was no significant increase up to 30 mins, indicating that the invasion process includes a lag-phase (Fig. 1). However, between 30 and 45 mins this number increased >10-fold to ~104 CFU (~0.1% inoculum). There was no further increase in the number of internalised S. aureus up to 90 minutes, indicating that maximal invasion levels had been reached (Fig. 1).
Figure 1. S. aureus adhesion to, and invasion of, keratinocyte and endothelial cells over time.
The total number of S. aureus 8325.4 (DU5883 ΔfnbA/B pFnBA4 expressing full length FnBPA) associated (adherent and internalized; circles) and internalized (squares) with each cell type was determined over a period of 90 minutes. Values for adhesion to, and invasion of, endothelial cells that are statistically significantly different from those obtained with HaCat cells are denoted (*). Experiments were performed 3 times in duplicate. Multiplicity of infection (MOI) = 20. Error bars represent the standard deviation of the mean.doi:10.1371/journal.pone.0018899.g001
To compare how invasion of keratinocytes compared to endothelial cells we also examined the adhesion to and invasion of EA. hy926 endothelial cells over the same time period (Fig. 1). S. aureus adhesion to endothelial cells was identical to that of keratinocytes after 15 minutes (Fig. 1). Adhesion to the endothelial cells increased over time and was significantly greater than HaCat cells after 45 mins. S. aureus invasion of endothelial cells was, however, 16-fold greater than that of keratinocytes after 15 minutes (Fig. 1). Endothelial cell invasion continued to increase until reaching saturation at 45–60 mins, at which point it was still approximately 6-fold greater than that seen with keratinocytes.
Low level keratinocyte invasion is not due to different invasion processes being utilised by S. aureus
Although several different pathogenic bacteria utilize integrins to trigger invasion, the cellular mechanisms involved can vary significantly. For example, Streptococcus pyogenes can invade via both caveolae and membrane ruffling depending on the invasin , . To examine whether the difference in efficiency of invasion of keratinocytes when compared to endothelial cells is due to the bacteria utilizing a different cellular process, we measured the internalisation of S. aureus FnBPR1–11 by HaCat cells pre-treated with inhibitors of cell function used previously in studies with endothelial cells . Disruption of host cell actin rearrangement (cytochalasin D), PI3K signalling (wortmannin), Src kinase signalling (PP2) or microtubule function (colchicine) all significantly inhibited S. aureus invasion, albeit to differing degrees (Fig. 2). By contrast, inhibition of tyrosine kinases (genistein), de novo protein synthesis (cycloheximide) or depletion of cholesterol (methyl-β-cyclodextrin) did not significantly affect invasion (Fig. 2). This was identical to that found when endothelial cells were studied , demonstrating that differences in these cellular process were not responsible for the difference in cell invasion efficiency observed.
Figure 2. Keratinocyte invasion occurs via a similar mechanism to endothelial cells.
HaCat cells were pre-incubated (60 min) with inhibitors of cell function before the addition of S. aureus FnBPAR1–11 (DU5883 ΔfnbA/B pFnBA4) and invasion determined after 90 mins. Inhibitors used were cytochalasin D (CD, inhibits actin polymerization), wortmannin (WRT, inhibits PI3-Kinase activity), genistein (GEN, inhibits tyrosine kinase activity), PP2 (PP2, Src kinase inhibitor), colchicine (COL, interferes with microtubule organisation), cycloheximide (CHX, inhibits eukaryotic protein synthesis) and methyl-β-cyclodextrine (MCD, depletes membrane cholesterol). Inhibitor-free medium was used as a positive control (CTL). Experiments were performed three times in duplicate. MOI = 20. Error bars represent the standard deviation of the mean. Values that are significantly different (p = <0.05) from control are indicated (*).doi:10.1371/journal.pone.0018899.g002
To confirm the role of cellular processes in invasion, we examined S. aureus-keratinocyte interactions using scanning electron microscopy (Fig. 3). After 15 minutes' incubation with S. aureus FnBPR1–11, there were significant numbers of attached bacteria but there were no visible changes in the host cell surface (Fig. 3A). After 30 minutes, a few adherent bacteria were associated with host-cell membrane alterations consistent with actin rearrangement (Fig. 3B–E). These morphological changes in the host cell membrane were consistent and we did not observe other types of changes, such as the formation of caveolae.
Figure 3. S. aureus FnBPR1–11 induces membrane ruffling and invasion of keratinocytes.
S. aureus FnBPR1–11 (DU5883 ΔfnbA/B pFnBA4) was incubated with keratinocytes for 15 (A), 30 (B–E) or 90 mins (F,G) before washing, fixation, processing and examination by scanning electron microscopy. As a negative control, S. aureus FnBPR0 was also examined after 90 mins (H). As a comparison, the interaction of S. aureus FnBPR1–11 with EA. Hy926 endothelial cells after 30 mins was also visualised (I–J). In each case the bar represents 1 µm. MOI = 20.doi:10.1371/journal.pone.0018899.g003
By 90 minutes post inoculation, a number of fully internalised bacteria could be observed within the keratinocytes (Fig. 3F,G). By contrast to S. aureus FnBPR1–11, S. aureus FnBPR0 did not induce any membrane alterations after 90 mins (Fig. 3H). We also examined the interaction of S. aureus FnBPR1–11 with EA. Hy926 cells. After 30 mins adherent bacteria were associated with membrane ruffles, similar to those observed with keratinocytes (Fig. 3I,J).
The composition of the FnBR region significantly affects invasion of keratinocytes
In a previous study we showed that a single high-affinity Fn binding repeat (FnBR) in FnBPA was sufficient for invasion of endothelial cells . To assess the role of FnBPA, and in particular the FnBR region, we examined adhesion to and invasion of HaCat cells by S. aureus expressing full length FnBPA: FnBPR1–11, a ΔfnbA/B mutant and S. aureus pFnBPR0 (which lacks the FnBR region) after 90 minutes (Fig. 4). Adhesion of S. aureus to HaCat cells was unaffected by the presence or absence of FnBPA (data not shown). Invasion, however, was dependent on the presence of FnBPA; S. aureus lacking FnBPs invaded at levels 15 times lower than S. aureus FnBPR1–11 (Fig. 5). FnBPA-triggered invasion was entirely dependent of the presence of the FnBR region; S. aureus FnBPR0 invaded HaCat cells at levels identical to S. aureus ΔfnbA/B (Fig. 5).
Figure 4. Diagrammatic representations of FnBPA and the variant constructs used in this study.
The relative positions of each of the major functional domains of FnBPA from S. aureus 8325.4 (SWISS-Prot P14738) (not to scale) are indicated, along with a schematic of each of the FnBPA variant constructs used in this study, which vary only in the composition of the FnBR domain. Low-affinity FnBRs are shaded.doi:10.1371/journal.pone.0018899.g004
Figure 5. The composition of the FnBR domain of FnBPA modulates host cell invasion.
Keratinocytes were incubated with either S. aureus DU5883 ΔfnbA/B (Δfnb) or S. aureus Δfnb expressing one of the FnBPA variant constructs detailed in figure 4 for 90 minutes and the number of internalized bacteria determined. Experiments were performed 3 times in duplicate. Error bars represent the standard deviation of the mean. MOI = 20. Values that are significantly different (p = <0.05) from S. aureus Δfnb are indicated (*).doi:10.1371/journal.pone.0018899.g005
To examine how the composition of the FnBR region modulates invasion, we measured internalisation of S. aureus expressing FnBPA variants containing various numbers of high- or low-affinity tandem repeats (Fig. 4). Previous work has demonstrated that these variants are expressed on the surface of S. aureus at equal levels and characterised the fibronectin-binding of each construct . Although S. aureus FnBPR1 or FnBPR11 invaded at levels significantly higher than S. aureus FnBPR0 this was >12-fold less than S. aureus FnBPR1–11 (Fig. 5). The presence of a second high affinity repeat (FnBPR10,11) was required to enhance internalisation to levels that were similar to that of S. aureus FnBPR1–11 (Fig. 5). Three high-affinity repeats (FnBPR1,10,11 and FnBPR9–11) also triggered invasion at similar levels to FnBPR1–11 (Fig. 5).
In contrast to the FnBPA variants with high-affinity repeats, S. aureus expressing FnBPA variants containing a single (FnBPR2 or FnBPR8) or two (FnBPR7,8) low-affinity repeats did not invade at levels that were significantly greater than FnBPR0 (Fig. 5). S. aureus expressing a FnBPA variant with three tandem low-affinity repeats (FnBPR6–8) did invade at levels above that of FnBPR0 but these were approximately 12-fold lower than S. aureus FnBPR1–11 (Fig. 5).
Over-expression of FnBPR1 or FnBPR6–8 fails to confer maximal invasion
We have shown previously that FnBPA-mediated invasion of endothelial cells is dependent on the total number of FnBRs on the bacterial cell surface, rather than the number within an individual FnBPA molecule . As such, high-level expression of FnBPR1 can trigger endothelial cell invasion to similar levels as S. aureus expressing FnBPR1–11. To determine whether maximal invasion of keratinocytes (FnBPR1–11 levels) could be triggered by over-expression of otherwise weakly-invasive FnBPR1 or FnBPR6–8 constructs, or whether invasion is absolutely dependent on multiple FnBRs within FnBPA, we employed a nisin-inducible system to control and over-express FnBPA variants on the surface of L. lactis . Previous work  has demonstrated that, at identical concentrations of nisin, expression levels of FnBPA variants on the surface of L. lactis are equal. L. lactis expressing each FnBPA construct at three surface expression levels (by inducing expression at three concentrations of nisin; 0, 10 or 100 ng ml−1) were assessed for invasion of HaCat cells (Fig. 6). At the lowest level of expression (0 ng ml−1 nisin, at which low-level expression occurs due to a leaky promoter) there were only negligible levels of invasion by L. lactis expressing each of the FnBPA variants (Fig. 6). Induction of FnBPR1–11 expression (10 ng ml−1 nisin) increased invasion 80-fold to levels similar to those seen with S. aureus FnBPR1–11 (Figures 1 and 5). Greater induction of FnBPR1–11 expression (100 ng ml−1) did not increase internalisation suggesting that maximal invasion levels had been reached (Fig. 6). Invasion of L. lactis FnBPR1,10,11 also increased dramatically when induced at 10 ng ml−1 nisin (relative to no induction), but at significantly lower levels than L. lactis FnBPR1–11 (Fig. 6). However, at the highest level of expression, L. lactis R1,10,11 was internalised at the same level as FnBPR1–11 (Fig. 6), mimicking the data obtained with S. aureus expressing FnBPA variants (Fig. 5). Induction of FnBPR1 or FnBPR6–8 expression (10 ng ml−1 nisin) promoted L. lactis invasion but at levels >30-fold lower than L. lactis FnBPR1–11 (Fig. 6). Further induction of FnBPR1 or FnBPR6–8 expression (100 ng ml−1 nisin) further increased invasion, but this was still 4–fold lower than that seen for L. lactis FnBPR1–11 at the same level of induction (Fig. 6).
Figure 6. Over-expression of FnBPR1 or FnBPR6–8 fails to trigger maximal invasion of Keratinocytes.
Expression of FnBPA variant constructs on the surface of L. lactis was performed using a nisin-inducible system. Invasion of L. lactis expressing FnBPAR1–11 (circles), FnBPAR1,10,11 (triangles), FnBPAR1 (squares) or FnBPAR6–8 (diamonds) was determined at 3 different levels of induction (0, 10 or 100 ng ml−1 nisin). Experiments were performed 3 times in duplicate. Error bars represent the standard deviation of the mean. MOI = 100. Values that are significantly different (p = <0.05) from L. lactis FnBPAR1–11 at identical nisin concentrations are indicated (*).doi:10.1371/journal.pone.0018899.g006
Inefficient invasion of keratinocytes may be due to weak α5β1 integrin expression
Although invasion processes do not seem to vary between endothelial cells and keratinocytes, a higher number of high affinity FnBRs are required for efficient invasion of keratinocytes. As such we hypothesised that the difference between the invasion efficiency of keratinocytes and endothelial cells may be due to differences in the density of the host cell ligand, the cell surface α5β1 integrin. To test this, we compared the expression of the α5 and β1 integrin subunits in keratinocytes and endothelial cells by performing a Western-immunoblot of whole-cell protein extracts. This revealed that α5β1 integrin expression levels were significantly higher in endothelial cells than keratinocytes (Fig. 7). This is consistent with studies of human skin , and may explain the differences in invasion levels between the two cells types.
Figure 7. Keratinocyte expression of α5 and β1 integrins is lower than endothelial cells.
HaCat keratinocytes (Kera) or EA. Hy926 endothelial (Endo) cells were harvested from flasks, lysed and the lysate examined by SDS-PAGE and Western-immunoblot to determine relative α5 or β1 integrin expression levels.doi:10.1371/journal.pone.0018899.g007
A FnBR peptide inhibits S. aureus invasion of keratinocytes
The rise in antibiotic resistance has prompted the search for novel approaches to preventing bacterial colonisation and infection. As there seems to be a critical ratio of FnBR and host ligand needed to invade cells ( Fig. 5 and 6), we hypothesized that an FnBR peptide might be sufficient to outcompete FnBPA and prevent S. aureus invasion. A tagless recombinantly-expressed peptide corresponding to FnBR9,10 (two high-affinity repeats) was used in an assay to examine its potential to block adhesion and invasion (Fig. 8). The peptide reduced adhesion approximately two-fold at the very highest concentration, despite FnBPA not being required for attachment of keratinocytes (Fig. 8, data not shown). By contrast, the peptide significantly inhibited S. aureus invasion at concentration of 1 nM (>2-fold reduction), which was even more pronounced at 2 nM (>5-fold) (Fig. 8). As such, this peptide may warrant further study as a potential novel prophylactic agent.
Figure 8. A recombinantly-expressed FnBR peptide inhibits S. aureus invasion.
The adhesion to (circles), and invasion (squares) of, keratinocytes by S. aureus FnBPR1–11 (DU5883 ΔfnbA/B pFnBA4) in the presence of various concentrations of recombinant FnBPA peptide (R9,10). Values that statistically significantly difference from those obtained in the absence of peptide are indicated (*). MOI = 20. Experiments were performed 3 times in duplicate. Error bars represent the standard deviation of the mean.doi:10.1371/journal.pone.0018899.g008
S. aureus is able to colonise and infect skin, causing a wide spectrum of infections ranging from superficial to invasive, acute to chronic . Infection can occur in healthy individuals as well as those with broken skin due to wounds, inserted medical devices or chronic skin conditions such as psoriasis . Central to colonisation and infection is the interaction of S. aureus with keratinocytes, which form an important barrier between the internal organs and external environment. Adhesion of S. aureus to keratinocytes results in inflammatory cytokine release and stimulates secretion of several antimicrobial peptides, of which β-defensin 3 reaches levels sufficient to kill S. aureus , . Cellular invasion is apparently not necessary to trigger these responses but internalised bacteria cause necrotic and apoptotic cell death . Interestingly, S. aureus adhesion to endothelial cells was equal to that of keratinocytes after 15 minutes, but was significantly greater after 90 minutes. This may reflect a decrease in keratinocyte-attached CFU due to the release of antimicrobial peptides by the HaCat cells.
Although the role of cellular invasion in colonisation and pathogenesis is unclear, previous work has shown that S. aureus can persist within keratinocytes for extended periods and it is possible that invasion provides shelter from the host antimicrobial arsenal. Intracellular S. aureus has been identified in chronic infections where it could form a protected reservoir as well as a mechanism of cellular dissemination and penetration of deeper tissues , .
We have shown previously that endothelial cell invasion is dependent upon the surface density of the FnBRs within FnBPA. The presence of multiple repeats within FnBPA is equivalent to multiple invasins in a single molecule, significantly enhancing efficiency and reducing immune exposure .
As the interaction of S. aureus with keratinocytes is a considerably more common event than with endothelial cells, we hypothesized that it likely provides the main selective pressure for the composition of FnBPs. The lack of a role for FnBPA in adhesion to HaCat keratinocyte cells is not surprising since S. aureus encodes a large number of adhesins. Indeed, at least five different S. aureus surface proteins are implicated in attachment to nasal epithelial cells, including ClfB, IsdA, SdrC, SdrD and SasG, as well as wall teichoic acid, which is essential for nasal colonisation –. By contrast, invasion of HaCat keratinocytes was FnBPA-dependent and, in keeping with previous work , the FnBR region was essential. Previous work has shown that a single high-affinity, or three low-affinity repeats were sufficient to trigger maximal invasion of endothelial cells . By contrast, our data indicate that keratinocyte invasion requires much higher surface densities of FnBRs than for invasion of endothelial cells. This may be a result of the significantly lower level of α5β1 integrin expression by the keratinocyte cells, which is in keeping with in vivo analysis of normal human skin . It is therefore possible that the difference in the speed of invasion is due to the increased length of time required for S. aureus to engage with sufficient α5β1 integrins on the keratinocyte cell surface and for subsequent cell-signalling events to occur. Additionally, the presence of multiple FnBRs within FnBPA may increase the efficiency of Fn binding through cooperative binding to arrays of FnBRs as observed recently for a Fn-binding protein from Streptococcus pyogenes . It is also possible that delayed entry of S. aureus into keratinocytes is desirable for the bacterium. A previous report  indicated that S. aureus delays uptake into endothelial cells in order to have sufficient time to prepare for intracellular life via up-regulation of e.g. toxin genes. It is possible that S. aureus employs a similar strategy for keratinocytes.
Although S. aureus invaded the endothelial cells more readily than the keratinocytes, the mechanism used appears to be the same (Figure 2 and ). Using identical concentrations of cell-function inhibitors used previously to study endothelial cell invasion , entry of S. aureus into HaCat cells was also found to involve PI3 and Src-kinases as well as actin rearrangement. Whilst one might expect genistein to have a similar effect to PP2, our data are not the first to suggest that genistein does not inhibit all the targets of PP2. Indeed, this phenomenon has been observed previously in studies of invasion mediated by fibronectin-binding protein invasins of S. pyogenes. Wang et al.  showed that invasion of epithelial cells by L. lactis expressing M-protein could be inhibited by PP2 but not by genistein. Conversely, invasion by L. lactis expressing the invasin SfbI was inhibited by genistein but not PP2. Interestingly, SfbI-mediated invasion relies on caveolae whilst M-protein mediated invasion does not, which ties in with our data showing no inhibition by methyl-β-cyclodextrin. It appears that, using our model system, FnBPA-mediated invasion is highly similar to that of S. pyogenes M-protein, rather than the structurally similar SfbI.
In summary, keratinocytes are significantly less amenable to invasion by S. aureus than endothelial cells, and require multiple repeats within FnBPA for invasion. As such, S. aureus interactions with keratinocytes and similar cells may provide the selection pressure for the multiple FnBRs within FnBPs. The poor penetration of many antibiotics into cells means that intracellular S. aureus could represent a reservoir for persistent infection , . Our data, in keeping with previous reports , , strongly suggest that FnBR peptides are highly effective at reducing S. aureus invasion and might form a novel prophylactic approach to reducing carriage and/or the development of chronic infections. Such an approach may have the added benefit of preventing invasion by other skin colonising pathogens such as S. pyogenes, which employ similar mechanisms of internalisation .
We thank Timothy Foster (Trinity College Dublin) for providing strains and the pFNBA4 vector. Garry Dunny (University of Minnesota) and Michiel Kleerebezem (Wageningen University) are thanked for the pMSP7517 vector and Howard Jenkinson (University of Bristol) for L. lactis NZ9800. HaCat keratinocytes were kindly provided by Angela Nobbs (University of Bristol).
Conceived and designed the experiments: AME JRP RCM. Performed the experiments: AME UP NAGM. Analyzed the data: AME UP NAGM JRP RCM. Contributed reagents/materials/analysis tools: AME UP NAGM JRP RCM. Wrote the paper: AME UP NAGM JRP RCM.
- 1. Lowy FD (1998) Staphylococcus aureus infections. N Engl J Med 339: 520–532.
- 2. David MZ, Daum RS (2010) Community-associated methicillin-resistant Staphylococcus aureus: epidemiology and clinical consequences of an emerging epidemic. Clin Microbiol Rev 23: 616–687.
- 3. Deleo FR, Otto M, Kreiswirth BN, Chambers HF (2010) Community-associated meticillin-resistant Staphylococcus aureus. Lancet 375: 1557–1568.
- 4. Chambers HF, Deleo FR (2009) Waves of resistance: Staphylococcus aureus in the antibiotic era. Nat Rev Microbiol 7: 629–641.
- 5. DeLeo FR, Chambers HF (2009) Reemergence of antibiotic-resistant Staphylococcus aureus in the genomics era. J Clin Invest 119: 2464–2474.
- 6. Kluytmans J, van Belkum A, Verbrugh H (1997) Nasal carriage of Staphylococcus aureus: epidemiology, underlying mechanisms, and associated risks. Clin Microbiol Rev 10: 505–520.
- 7. Chiller K, Selkin BA, Murakawa GJ (2001) Skin microflora and bacterial infections of the skin. J Investig Dermatol Symp Proc 6: 170–174.
- 8. Brook I (2002) Secondary bacterial infections complicating skin lesions. J Med Microbiol 51: 808–812.
- 9. Lowy FD (2000) Is Staphylococcus aureus an intracellular pathogen? Trends Microbiol 8: 341–343.
- 10. Clement S, Vaudaux P, Francois P, Schrenzel J, Huggler E, et al. (2005) Evidence of an intracellular reservoir in the nasal mucosa of patients with recurrent Staphylococcus aureus rhinosinusitis. J Infect Dis 192: 1023–1028.
- 11. Sinha B, Herrmann M (2005) Mechanism and consequences of invasion of endothelial cells by Staphylococcus aureus. Thromb Haemost 94: 266–277.
- 12. Plouin-Gaudon I, Clement S, Huggler E, Chaponnier C, François P, et al. (2006) Intracellular residency is frequently associated with recurrent Staphylococcus aureus rhinosinusitis. Rhinology 44: 249–254.
- 13. von Eiff C, Peters G, Becker K (2006) The small colony variant (SCV) concept – the role of staphylococcal SCVs in persistent infections. Injury 37: S26–33.
- 14. Garzoni C, Kelley WL (2009) Staphylococcus aureus: new evidence for intracellular persistence. Trends Microbiol 17: 59–65.
- 15. Sendi P, Proctor RA (2009) Staphylococcus aureus as an intracellular pathogen: the role of small colony variants. Trends Microbiol 17: 54–58.
- 16. Zautner AE, Krause M, Stropahl G, Holtfreter S, Frickmann H, et al. (2010) Intracellular persisting Staphylococcus aureus is the major pathogen in recurrent tonsillitis. PLoS One 5: e9452.
- 17. Edwards AM, Potts JR, Josefsson E, Massey RC (2010) Staphylococcus aureus host cell invasion and virulence in sepsis is facilitated by the multiple repeats within FnBPA. PLoS Pathog 6: e1000964.
- 18. Sinha B, François PP, Nüsse O, Foti M, Hartford OM, Vaudaux P, et al. (1999) Fibronectin-binding protein acts as Staphylococcus aureus invasin via fibronectin bridging to integrin α5β1. Cell Microbiol 1: 101–117.
- 19. Sinha B, Francois P, Que YA, Hussain M, Heilmann C, et al. (2000) Heterologously expressed Staphylococcus aureus fibronectin-binding proteins are sufficient for invasion of host cells. Infect Immun 68: 6871–6878.
- 20. Fowler T, Wann ER, Joh D, Johansson S, Foster TJ, Höök 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: 672–629.
- 21. Nitsche-Schmitz DP, Rohde M, Chhatwal GS (2007) Invasion mechanisms of Gram-positive pathogenic cocci. Thomb Haemost 98: 488–496.
- 22. Keane FM, Loughman A, Valtulina V, Brennan M, Speziale P, Foster TJ (2007) Fibrinogen and elastin bind to the same region within the A domain of fibronectin binding protein A, an MSCRAMM of Staphylococcus aureus. Mol Microbiol 63: 711–723.
- 23. O'Neill E, Pozzi C, Houston P, Humphreys H, Robinson DA, et al. (2008) A novel Staphylococcus aureus biofilm phenotype mediated by the fibronectin-binding proteins, FnBPA and FnBPB. J Bacteriol 190: 3835–3850.
- 24. Meenan NA, Visai L, Valtulina V, Schwarz-Linek U, Norris NC, et al. (2007) The tandem beta-zipper model defines high affinity fibronectin-binding repeats within Staphylococcus aureus FnBPA. J Biol Chem 282: 25893–25902.
- 25. Fröman G, Switalski LM, Speziale P, Höök MJ (1987) Isolation and characterization of a fibronectin receptor from Staphylococcus aureus. Biol Chem 262: 6564–6571.
- 26. Ingham KC, Brew S, Vaz D, Sauder DN, McGavin MJ (2004) Interaction of Staphylococcus aureus fibronectin-binding protein with fibronectin: affinity, stoichiometry, and modular requirements. J Biol Chem 279: 42945–42953.
- 27. Bingham RJ, Rudiño-Piñera E, Meenan NA, Schwarz-Linek U, Turkenburg JP, et al. (2008) Crystal structures of fibronectin-binding sites from Staphylococcus aureus FnBPA in complex with Fn domains. Proc Natl Acad Sci U S A 105: 12254–12258.
- 28. Ozeri V, Rosenshine I, Ben-Ze'Ev A, Bokoch GM, Jou TS, Hanski E (2001) De novo formation of focal complex-like structures in host cells by invading Streptococci. Mol Microbiol 41: 561–573.
- 29. Schwarz-Linek U, Höök M, Potts JR (2006) Fibronectin-binding proteins of gram-positive cocci. Microbes Infect 8: 2291–2298.
- 30. Massey RC, Kantzanou MN, Fowler T, Day NP, Schofield K, et al. (2001) Fibronectin-binding protein A of Staphylococcus aureus has multiple, substituting, binding regions that mediate adherence to Fn and invasion of endothelial cells. Cell Microbiol 3: 839–851.
- 31. Peacock SJ, Day NP, Thomas MG, Berendt AR, Foster TJ (2000) Clinical isolates of Staphylococcus aureus exhibit diversity in fnb genes and adhesion to human fibronectin. J Infect 41: 23–31.
- 32. von Eiff C, Becker K, Metze D, Lubritz G, Hockmann J, et al. (2001) Intracellular persistence of Staphylococcus aureus small-colony variants within keratinocytes: a cause for antibiotic treatment failure in a patient with darier's disease. Clin Infect Dis 32: 1643–1647.
- 33. Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory press.
- 34. Boukamp P, Dzarlieva-Petrusevska RT, Breitkreuz D, Hornung J, Markham A, Fusenig NE (1988) Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte cell line. J Cell Biol 106: 761–771.
- 35. Boukamp P, Popp S, Altmeyer S, Hülsen A, Fasching , et al. (1997) Sustained nontumorigenic phenotype correlates with a largely stable chromosome content during long-term culture of the human keratinocyte line HaCat. Gene Chromosome Canc 19: 201–214.
- 36. Mempel M, Schmidt T, Weidinger S, Schnopp C, Foster T, et al. (1998) Role of Staphylococcus aureus surface-associated proteins in the attachment to cultured HaCaT keratinocytes in a new adhesion assay. J Invest Dermatol 111: 452–456.
- 37. Mempel M, Schnopp C, Hojka M, Fesq H, Weidinger S, et al. (2002) Invasion of human keratinocytes by Staphylococcus aureus and intracellular bacterial persistence represent haemolysin-independent virulence mechanisms that are followed by features of necrotic and apoptotic keratinocyte cell death. Br J Dermatol 146: 943–951.
- 38. Deyrieux AF, Wilson VG (2007) In vitro culture conditions to study keratinocyte differentiation using the HaCaT cell line. Cytotechnology 54: 77–83.
- 39. Dombek PE, Cue D, Sedgewick J, Lam H, Finlay BB, Cleary PP (1999) High-frequency intracellular invasion of epithelial cells by serotype M1 group A streptococci: M1 protein-mediated invasion and cytoskeletal rearrangements. Mol Microbiol 31: 859–870.
- 40. Rohde M, Muller E, Chhatwal GS, Talay SR (2003) Host cell caveolae act as an entry-port for group A streptococci. Cell Microbiol 5: 323–342.
- 41. Pellegrini G, De Luca M, Orecchia G, Balzac F, Cremona O, et al. (1992) Expression, topography, and function of integrin receptors are severely altered in keratinocytes from involved and uninvolved psoriatic skin. J Clin Invest 89: 1783–1795.
- 42. Aufiero B, Guo M, Young C, Duanmu Z, Talwar H, et al. (2007) Staphylococcus aureus induces the expression of tumor necrosis factor-alpha in primary human keratinocytes. Int J Dermatol 46: 687–694.
- 43. Kisich KO, Howell MD, Boguniewicz M, Heizer HR, Watson NU, Leung DY (2007) The constitutive capacity of human keratinocytes to kill Staphylococcus aureus is dependent on beta-defensin 3. J Invest Dermatol 127: 2368–2380.
- 44. Roche FM, Meehan M, Foster TJ (2003) The Staphylococcus aureus surface protein SasG and its homologues promote bacterial adherence to human desquamated nasal epithelial cells. Microbiology 149: 2759–2767.
- 45. Corrigan RM, Miajlovic H, Foster TJ (2009) Surface proteins that promote adherence of Staphylococcus aureus to human desquamated nasal epithelial cells. BMC Microbiol 30: 22.
- 46. Weidenmaier C, Kokai-Kun JF, Kristian SA, Chanturiya T, Kalbacher H, et al. (2004) Role of teichoic acids in Staphylococcus aureus nasal colonization, a major risk factor in nosocomial infections. Nat Med 10: 243–245.
- 47. Clarke SR, Brummell KJ, Horsburgh MJ, McDowell PW, Mohamad SA, et al. (2006) Identification of in vivo-expressed antigens of Staphylococcus aureus and their use in vaccinations for protection against nasal carriage. J Infect Dis 193: 1098–1108.
- 48. Marjenberg ZR, Ellis IR, Hagan RM, Prabhakaran S, Hook M, et al. (2010) Cooperative binding and activation of fibronectin by a bacterial surface protein. J Biol Chem 286: 1884–1894.
- 49. Schröder A, Schröder B, Roppenser B, Linder S, Sinha B, et al. (2006) Staphylococcus aureus fibronectin binding protein-A induces motile attachment sites and complex actin remodeling in living endothelial cells. Mol Biol Cell 17: 5198–5210.
- 50. Wang B, Li S, Dedhar S, Cleary PP (2007) Paxillin phosphorylation: bifurcation point downstream of integrin-linked kinase (ILK) in streptococcal invasion. Cell microbial 9: 1519–1528.
- 51. Hirt H, Erlandsen SL, Dunny GM (2000) Heterologous inducible expression of Enterococcus faecalis pCF10 aggregation substance asc10 in Lactococcus lactis and Streptococcus gordonii contributes to cell hydrophobicity and adhesion to fibrin. J Bacteriol 182: 2299–2306.
- 52. Greene C, McDevitt D, Francois P, Vaudaux PE, Lew DP, Foster TJ (1995) Adhesion properties of mutants of Staphylococcus aureus defective in fibronectin-binding proteins and studies on the expression of fnb genes. Mol Microbiol 17: 1143–1152.
- 53. Kuipers OP, Beerthuyzen MM, Siezen RJ, De Vos WM (1993) Characterization of the nisin gene cluster nisABTCIPR of Lactococcus lactis. Requirement of expression of the nisA and nisI genes for development of immunity. Eur J Biochem 216: 281–291.