The Staphylococcus aureus Lipoprotein SitC Colocalizes with Toll-Like Receptor 2 (TLR2) in Murine Keratinocytes and Elicits Intracellular TLR2 Accumulation ^{▿ †}

Bacterial strains, growth media, and plasmids.

S. aureus SA113, S. aureus(pTX30SitC-his), and Staphylococcus carnosus ΔRKET(pPSHG5ΩHis6mSHL) were used in this study. Strains were grown aerobically at 37°C in basal medium (BM) with the supplements indicated in Table 1.

Mouse strains.

C57BL/6 wild-type mice were purchased from Charles River (Sulzfield, Germany), and Nod2-deficient mice were purchased from Jackson Laboratory, ME. TLR2-deficient mice were obtained from C. Kirschning (Technical University, Munich, Germany). All deficient strains were in the C57BL/6 background. The animals were bred under specific-pathogen-free conditions at the animal facility of the University of Tübingen according to European guidelines and German law, and the preparation of oral mucosa was approved by the Regierungspräsidium Tübingen (Az 25.04.07 and Az 05.06.09).

Cells and cell lines.

Primary cultures of murine epithelial cells (keratinocytes) were obtained from adult oral mucosa. After overnight treatment with the epidermis upside down in trypsin solution at 4°C, the epidermis was separated from the dermis, and epidermal cells were collected by centrifugation. Murine epithelial cells were cultured in defined medium (64.5% Dulbecco's modified Eagle's medium [DMEM], 21.5% Ham's medium, 10% fetal calf serum, 2% penicillin-streptomycin) for 2 days at 37°C and 5% CO₂. Subsequently, cells were transferred to a second medium (94% MCDB [basal nutrient medium], 2% fetal calf serum, 2% penicillin-streptomycin) and grown for 3 days before experiments were started. Human Mono Mac 6 (MM6) cells were cultured in RPMI 1640 medium (Biochrom AG, Berlin, Germany) supplemented with 10% fetal calf serum (Biochrom AG, Berlin, Germany), 2× nonessential amino acids, and OPI supplement (Sigma-Aldrich, Taufkirchen, Germany). Cells were grown for 3 days before experiments were started.

Construction of SitC-His-expressing plasmid pTX30SitC-his.

The sitC gene was cloned into a derivative of the xylose-inducible expression vector pTX15 (37) and expressed with a C-terminal extension of 12 amino acids (thrombin cleavage site and histidine tag [KVPRGSHHHHHH]), and the product was named SitC-His (Fig. 1 A). sitC was amplified from S. aureus SA113 genomic DNA by using the primer pair BamHI-fwd (TATTTAGGATCCGAAACGAGGAAGTTTAACATGAAAAAATTAG) and SacI-rev (ATAATTGAGCTCTTAATGATGATGATGATGATGTGAACCACGTGGAACTAATTTCAG CTTCCGTGTAC) (underlining indicates restriction enzyme sites) and High Fidelity PCR enzyme mix (Fermentas, St. Leon-Rot, Germany). The sequence of the PCR product (999 bp) contained the sitC gene with a C-terminal thrombin-cleavable His₆ tag. This PCR product was cloned in frame into the polylinker upstream of the xylA promoter region of the pTX15 derivative, resulting in pTX30sitC-His. This plasmid was transformed into S. aureus SA113 by electroporation (5), and the insert was sequenced using a Li-Cor Long Reader DNA sequencer (Lincoln Corporation, Inc., Lincoln, NE). Computer sequence analysis was performed with the Vector NTI program.

Purification of SitC-His.

S. aureus(pTX30SitC-his) cells were grown at 37°C in the presence of xylose for 13 h prior to growth at 37°C in the absence of xylose for 2 h. Cells were harvested and broken via FastPrep. Cell membranes were obtained by ultracentrifugation. SitC-His was solubilized in extraction buffer (20 mM Tris-HCl, 50 mM NaCl, 2% Triton X-100, pH 8.0) for 18 h at 6°C. After another ultracentrifugation step, the supernatant was taken and incubated with Ni-nitrilotriacetic acid (Ni-NTA) superflow beads (Qiagen, Hilden, Germany) that had been equilibrated with extraction buffer for 18 h at 6°C. The beads were then used to fill a column and washed four times with washing buffer (20 mM Tris-HCl, 50 mM NaCl, 0.25% Triton X-100, 40 mM imidazole) prior to elution with 20 mM Tris-HCl, 50 mM NaCl, 0.25% Triton X-100, and 400 mM imidazole, pH 8.0. The elution was performed in two steps. The two fractions were combined and concentrated 10-fold via a Vivaspin centrifugal ultrafilter unit with a molecular mass cutoff of 10 kDa (Sartorius, Göttingen, Germany). The concentrated eluate was analyzed by SDS-PAGE (Fig. 2 A). The purified protein was concentrated, and the total amount was estimated by Bradford assay and stored at −80°C.

Purification of SHL from S. carnosus ΔRKET(pPSHG5ΩHis6mSHL).

Mature lipase (without PP) with an N-terminal His tag (SHL) was expressed intracellularly in S. carnosus ΔRKET(pPSHG5ΩHis-SHL) (26). SHL was isolated under denaturing conditions with 8 M urea and subsequently purified by Äkta fast-performance liquid chromatography (FPLC) with a His-trap column.

Determination of LPS contamination.

Lipopolysaccharide (LPS) concentrations in the stock solutions of the stimulants used for this study were determined by QCL-1000 LAL assay (Cambrex, Walkersville, MD). The LPS concentrations in 10-μg preparations of SitC-His, Pam₃Cys, and SHL were 0.005 endotoxin unit (EU), 0.02 EU, and 0.01 EU, respectively (10 EU = 1 ng LPS).

Labeling of purified proteins.

Protein buffers were changed to LPS-free 0.1 M sodium carbonate buffer, pH 10 to 11. Cy2 or Cy5 labeling was performed as recommended by the manufacturer (GE Healthcare, Freiburg, Germany). The samples were spun at 5,000 × g for 90 min at room temperature (RT) in a centrifugal ultrafilter unit (cutoff, 10 kDa) to remove unbound dye and to change the buffer conditions to neutral pH (LPS-free phosphate-buffered saline [PBS]). Fluorescently labeled proteins were finally diluted in 1 ml LPS-free PBS and stored at −80°C.

Reagents used in cell culture.

Penicillin-streptomycin, Accutase, nonessential amino acids, and PBS were obtained from PAA (Pasching, Austria). LPS from E. coli was purchased from Sigma-Aldrich (Taufkirchen, Germany). The synthetic lipopeptide Pam₃Cys (EMC Microcollections, Tübingen, Germany), mirroring the diacylated lipoproteins of bacteria, is also known to activate TLR2 (7). The antibodies used were rabbit anti-TLR2, rabbit anti-TLR4, rabbit anti-Nod2 (all from Santa Cruz Biotechnology, Santa Cruz, CA), and a Cy3-conjugated donkey anti-rabbit antibody (Jackson ImmunoResearch Laboratory, West Grove, PA).

Sample preparation for confocal microscopy.

A total of 3 × 10⁵ freshly isolated, live keratinocytes were seeded in chamber slides for 3 to 5 days at 37°C, 5% CO₂, and 95% relative humidity. Afterwards, cells were incubated with different stimuli for 15 to 180 min. Cells were fixed with PLP (0.1 M l-lysin-HCl, 2% paraformaldehyde, 0.01 M sodium meta-periodate, pH 7.4) for 10 min at RT and permeabilized with 0.5% Triton X-100 for 10 min at RT. Cells were subsequently stained with antibodies against TLR2, Nod2, and TLR4 (all from rabbits). A Cy3-conjugated anti-rabbit antibody was used as a secondary antibody. Nuclei were stained with DAPI (4′,6-diamidino-2-phenylindole; Invitrogen, Karlsruhe, Germany) or Yo-Pro-1 iodide (Invitrogen, Karlsruhe, Germany). Cells were observed by confocal laser scanning microscopy (CLSM), using a Leica TCS SP 2 spectral confocal and multiphoton inverted microscope (Leica, Heidelberg, Germany). Images were taken using a Plan-Apochromat 63×/1.32-numerical-aperture oil immersion objective, with fluorescence excitation at 488 nm (Ar laser), 543 nm (HeNe laser), and 633 nm (HeNe laser) and with an Enterprise II 351/364-nm DAPI laser (Coherent, Dieburg, Germany). Appropriate filters were selected for individual staining. Fluorescence analysis was carried out by Leica confocal software.

Cytokine enzyme-linked immunosorbent assay (ELISA).

For stimulation experiments, Mono Mac 6 cells were washed two times with PBS and seeded at 500 μl/well in 24-well plates at a density of 10⁶ cells/ml. Mono Mac 6 cells were pulsed with each SitC-His protein as well as with Pam₃Cys for 24 h at 37°C in a humidified 5% CO₂ environment. MKs were seeded at 500 μl/well in 24-well plates at a density of 10⁵ cells/ml. Cells were cultured in defined medium for 2 days at 37°C and 5% CO₂. Subsequently, cells were transferred to a second medium and grown for 3 days before experiments were started. MKs were stimulated with different amounts of SitC-His, Pam₃Cys, and LPS for 48 h.

The cell supernatants were collected and stored at −20°C. Human interleukin-6 (hIL-6), human tumor necrosis factor alpha (hTNF-α), and murine IL-6 (mIL-6) levels were determined by use of ELISA kits according to the manufacturer's instructions (eBioscience, San Diego, CA). The optical density was measured at 450 nm and 570 nm by a Tecan Infinite 200 reader (Tecan, Crailsheim, Germany).

Reporter assay of NF-κB activation.

Human embryonic kidney (HEK293) cells (ATCC, Manassas, VA) were cultured in DMEM supplemented with 2 mM l-glutamine and 10% fetal calf serum (all from Biochrom, Berlin, Germany). HEK293 and stably transfected HEK293-hTLR2 cells (Invivogen, San Diego, CA) were plated at 2 × 10⁵ cells/well in 24-well plates 1 day before transfection. Cells were then transiently transfected by use of Lipofectamine 2000 transfection reagent (Invivogen, San Diego, CA) with 100 ng of an NF-κB reporter plasmid (pNFkB-TA-Luc; Clontech, BD Biosciences, Heidelberg, Germany) and 10 ng of a plasmid that directed expression of Renilla luciferase under the control of the constitutively active thymidine kinase promoter (pRL-TK; Promega). The pcDNA3.1 vector (Promega) was used to balance the transfected DNA concentration. Twenty-four hours after transfection, cells were stimulated with cell wall components in serum-free medium for 24 h, and luciferase activity was measured using a dual-luciferase reporter assay system (Promega) according to the manufacturer's instructions.

Cy2-SitC-His incorporates into MKs and elicits TLR2 accumulation.

Although it was recently described that the S. aureus lipoprotein SitC has an immune-stimulatory activity toward TLR2 (27), internalization of SitC into host cells and colocalization with TLR2 have not yet been studied. To monitor the effect of SitC on TLR2 synthesis in MKs, these cells were stimulated with different amounts of Cy2-SitC-His (15, 20, and 40 μg, corresponding to 0.45, 0.6, and 1.2 nmol, respectively) over different periods (15, 30, 90, and 180 min) and analyzed by CLSM. As controls, the synthetic lipopeptide Pam₃Cys and purified Cy5-SHL lipase were used. Cy2-SitC-His stimulated intracellular TLR2 production which was time (15, 30, and 90 min) and concentration (15 and 40 μg of Cy2-SitC-His) dependent (Fig. 3 and 4).

In nonstimulated MKs (PBS controls), no TLR2 was detectable (Fig. 3A). Stimulation of MKs with the synthetic lipopeptide Pam₃Cys (Fig. 3B), which mimics triacylated lipoproteins, and with nonlabeled SitC-His (Fig. 3C) augmented intracellular TLR2 accumulation. Cy2-SitC-His was internalized into MKs (Fig. 3D and E) and elicited intracellular production and accumulation of TLR2 (Fig. 3F). After 90 min of stimulation, the TLR2 accumulation was significantly increased in comparison to that at 15 and 30 min of stimulation (Fig. 4A and B). Stimulation for over 90 min did not lead to further increases of detectable intracellular Cy2-SitC-His and TLR2 (data not shown). Furthermore, stimulation with 40 μg of Cy2-SitC-His induced a significantly stronger accumulation of TLR2 than did stimulation with 20 μg for the same amount of time (Fig. 4B and C).

Cy2-SitC-His was taken up by TLR2-deficient MKs, but TLR2 was not detected (Fig. 3D). Thus, the cell uptake of Cy2-SitC-His appeared not to be TLR2 dependent. Uptake of Cy2-SitC-His did not lead to Nod2 augmentation in MKs (Fig. 3F). Nod2 was detectable after stimulation with muramyl dipeptides (Fig. 3G).

Cy2-SitC-His did not affect TLR4 (Fig. 5 A to C), a known LPS receptor (9, 17). These data demonstrate the specificity of our results. The occurrence of TLR2 was not enhanced after stimulation with Cy5-SHL, which contains no lipid modification (Fig. 5D).

SitC-His colocalizes with TLR2 and mediates TLR2-dependent signaling.

Cy2-SitC-His not only induced intracellular TLR2 accumulation but also was internalized and colocalized with TLR2 (Fig. 3). This colocalization indicates an interaction of the SitC lipoprotein with TLR2, which is thought to be essential for triggering the innate immune response.

Staphylococcal SitC has been shown to induce a TLR2-dependent activation of the innate immune system (27, 43). This was confirmed here with an NF-κB luciferase reporter plasmid in TLR2- and/or Nod2-expressing HEK293 cells. Only TLR2-expressing HEK cells showed an increase in luciferase expression when stimulated with SitC-His or Pam₃Cys; no stimulation was seen with LPS (Fig. 6). Stimulation with SitC-His resulted in proinflammatory cytokine production (TNF-α and IL-6), as investigated with the human monocytic cell line Mono Mac 6 (Fig. 7 A and B). SitC-His also induced the release of murine IL-6 in wild-type MKs, Nod2-deficient MKs, and TLR4-deficient MKs but not in TLR2-deficient MKs (Fig. 7C).

Taken together, these data demonstrate that the S. aureus lipoprotein SitC-His is incorporated into MKs, provokes intracellular TLR2 accumulation, colocalizes with TLR2, and induces a TLR2-dependent immune response.