Significance of the d-Serine-Deaminase and d-Serine Metabolism of Staphylococcus saprophyticus for Virulence

Bacterial strains and plasmids.The bacterial strains and plasmids used in this study are listed in Table 1. S. saprophyticus strain 7108, a hemagglutinating, fibronectin- and collagen-binding clinical isolate has been described previously (33–35). E coli DH5α (36) was the host for expression experiments and the intermediate host during construction of the plasmids for allelic replacement. The shuttle plasmid pBT2 (37) contains the temperature-sensitive replicon of pE194, the chloramphenicol resistance of pC194 and the multiple cloning site of pUC18. Plasmid pEC1 (38) was used as the erm(B) source. The pPS44 vector (39) was used for cloning experiments involving Staphylococcus carnosus strain TM300 (40) and for complementation experiments. The plasmid pMB2200, conferring tetracycline resistance, had been isolated from a clinical isolate of S. saprophyticus (41).

Bacterial growth media and antibiotics.E. coli strains harboring plasmids were grown in lysogeny broth (LB) or on LB agar plates. S. saprophyticus strain 7108 was grown in peptone-yeast extract (PY) broth or tryptic soy broth (TSB) (Oxoid, Wesel, Germany) or on agar plates. Bacteria were usually incubated at 37°C, but in some experiments, temperatures of 30°C and 42°C were also used. Ampicillin (100 μg/ml) was used for the selection of plasmids in E. coli. For selection of plasmids or chromosomal markers in S. saprophyticus, 10 μg/ml chloramphenicol, 5 μg/ml erythromycin, or 10 μg/ml tetracycline was used.

DNA preparation.Plasmid DNA was prepared using the Qiagen plasmid Midi kit (Qiagen Hilden, Germany) and genomic DNA with the Qiagen DNeasy kit (Qiagen, Hilden, Germany). For DNA preparations from staphylococci, pelleted bacteria were resuspended in lysis buffer (20 mM Tris [pH 8.0], 2 mM EDTA, 1.2% [vol/vol] Triton X-100). To lyse the bacteria, we added 100 μg/ml Amibicin L (Wak Chemie Medici GmbH, Steinbach, Germany) to the lysis buffer.

Construction of an insertion mutant by allelic replacement.The dsdA gene was deleted by the insertion of the erm(B) resistance gene, and pBT2 (37) was used as the replacement vector. To this end, a 700-bp PCR amplificate of the chromosomal DNA upstream of the dsdA gene generated with the primers Ssa.dsd1seq/BamHI and Ssa.dsd2/2rev/XbaI (Table 2) was digested with BamHI and XbaI and was cloned into pUC18. This plasmid was designated pMB1401. For the construction of pMB1402, we inserted the erm(B) cassette cut from pEC1 into plasmid pMB1401 digested with the enzymes XbaI and PstI. A further 600-bp DNA amplificate downstream of the dsdA gene was generated with the primers Ssa.dsd1rev/HindIII and Ssa.dsd2seq/PstI and was ligated into pMB1402 cut with these enzymes. The resultant plasmid (pMB1403) contained the upstream and downstream flanking regions of the dsdA gene with an erm(B) cassette in the center. The insert of pMB1403 was excised with BamHI and HindIII and ligated into the temperature-sensitive replacement vector pBT2, yielding pMB1404.

The pMB1404 constructs were purified from E. coli DH5α and transformed into S. saprophyticus strain 7108 by protoplast transformation, as previously described (12). Chloramphenicol- and erythromycin-resistant clones were grown in the presence of erythromycin (5 μg/ml at 30°C for 24 h) and used to inoculate 1,000 ml of prewarmed (42°C) broth containing erythromycin. After overnight incubation, appropriate dilutions were plated onto P-agar containing erythromycin. Clones that grew on erythromycin but not on chloramphenicol had lost the plasmid, and correct insertion of the erm(B) cassette was checked by PCR and sequencing. Absence of the cat gene was verified by PCR.

Complementation of the ΔdsdA mutant.The vector part of pPS44 (39) was amplified by inverse PCR with the primers pPS44BamHI and pPS44HindIII. The dsdA gene with its own promoter was amplified with the primers Ssa.dsdA4seq and Ssa.dsdA4rev, and both were digested with BamHI and HindIII. Ligation and transformation of S. carnosus TM300 (40) were performed as described elsewhere (42). The resulting plasmid (pMB1406) was purified from this strain and introduced into the ΔdsdA mutant by protoplast transformation (12).

Animal experiments.Eight-week-old female C3H/HeN mice (Harlan) were infected by transurethral catheterization as previously described (43). Static bacterial cultures were started from freezer stocks, grown at 37°C for 18 h in brain heart infusion (BHI) broth and then subcultured at 1:250 or 1:100 into fresh medium. These subcultures were then grown statically at 37°C for 18 h, pelleted, and resuspended in phosphate-buffered saline (PBS), to yield 50-μl inocula containing 1 × 10⁷ to 2 × 10⁷ CFU. For competition experiments, bacteria were diluted appropriately to achieve 10⁷ CFU each of the wild type and dsdA mutant together in a total volume of 50 μl. To quantify bacteria present in mouse organs, bladder and kidneys were aseptically harvested at the indicated times postinfection, homogenized in phosphate-buffered saline, serially diluted, and plated onto BHI agar plates for total CFU enumeration and with erythromycin for mutant CFU (CFU_m) enumeration. Wild-type CFU (CFU_w) were calculated as total CFU − mutant CFU. In samples where zero CFU were recovered, the limit of detection of the protocol was used. Competition indices (CI) were calculated similarly to the method described by Freter et al. (44) by using the wild type as the reference strain, as follows: CI = (CFU_m/CFU_w recovered from mice)/(CFU_m/CFU_w present in the initial inoculum).

Significance was calculated using the nonparametric Wilcoxon signed-rank test (two tailed) comparing the theoretical median CI to 1 (GraphPad Prism 5). All animal studies were performed in accordance with the Committee for Animal Studies at Washington University School of Medicine.

Cocultivation experiments.Cocultivation experiments were carried out in lysogeny broth (LB) (Invitrogen, Karlsruhe, Germany) supplemented with 2.5, 5, or 10 mg/ml d-serine (AppliChem, Darmstadt, Germany) or l-serine (AppliChem) or without any supplements. Bacteria were precultured overnight in 10 ml medium at 37°C with a flask-to-medium ratio of 100:10 and with shaking at 100 rpm. Cells from overnight cultures were then resuspended in 0.9% NaCl, and the optical density at 600 nm (OD₆₀₀) was adjusted to 0.5. For inoculation of the coculture, 100 μl of each strain was added to 10 ml of medium. Cultures were then incubated for 24 h at 37°C and 100 rpm. Every 24 h, the mixed populations were diluted 1:100 into fresh medium. Growth was checked by determining CFU every 48 h by serial dilutions (10⁻³ to 10⁻⁹ in 0.9% NaCl). Each dilution was plated onto three LB agar plates (LB agar powder; AppliChem): one without and two with different antibiotics for differentiation of the strains. Erythromycin (10 μg/ml; Sigma-Aldrich) was used for selection of the mutant, chloramphenicol (20 μg/ml; AppliChem) for selection of the complemented mutant, and tetracycline (10 μg/ml; AppliChem) for positive selection of the wild type, which had been transformed with a tetracycline resistance plasmid (pMB2200) isolated from a clinical isolate of S. saprophyticus. Agar plates were incubated overnight at 37°C, and bacterial counts were determined.

Chemically defined medium.Chemically defined medium contained the following (per liter of water): 7 g K₂HPO₄ · 3H₂O, 2 g KH₂PO₄, 0.4 g Na₃ citrate · 2H₂O, 0.05 g MgSO₄, 1 g (NH₄)₂SO₄, 20 mg thymine, 0.5 g sodium thiosulfate, 50 mg l-arginine, 100 mg l-glutamine, 50 mg l-glycine, 20 mg l-histidine, 90 mg l-leucine, 50 mg l-lysine, 3 mg l-methionine, 40 mg l-phenylalanine, 80 mg l-proline, 30 mg l-threonine, 10 mg l-tryptophan, 80 mg l-valine, 5 mg FeSO₄ · 7H₂O, 10 mg CaCl₂, 7 mg ZnCl₂, 13 mg NiSO₄ · 6H₂O, 10 mg MnCl₂ · 4H₂O, 0.5 mg thiamine hydrochloride, 0.6 mg nicotinic acid, 0.125 mg d-pantothenic acid calcium salt, and 0.125 mg biotin, adjusted to pH 7.0. Before use, the medium was filter sterilized. To exclude contaminating constituents from residues in reusable glassware, only new plastic materials were used for preparation and growth. Also, bacterial cultures were grown in plastic tubes (10-ml culture in a 50-ml Falcon tube). Without addition of the carbon and energy source (glucose or d-serine), bacteria were not able to grow.

Pyruvate assay.The assay was conducted basically as described by McFall (45). It measures the concentration of a colored α-keto acid that is formed as the product of a reaction of pyruvate with 2,4-dinitrophenylhydrazine under alkaline conditions. The strains used in this assay were cultured in 100 ml peptone-yeast extract (PY) broth (10 g/liter peptone, 5 g/liter yeast, 1 g/liter glucose, 5 g/liter NaCl, 1.25 g/liter Na₂HPO₄ · 2H₂O [pH 7.3]) overnight (37°C and 100 rpm). The medium was supplemented with d-serine (1 g/liter) to ensure that the d-serine-deaminase was expressed. Twenty-five milliliters of the overnight culture was pelleted by centrifugation at 4,000 rpm for 10 min. The bacteria were then washed with 0.07 M PBS (pH 7.4) and resuspended in 2 ml of this buffer. The following lysis was done by addition of 10 μl lysostaphin (5 mg/ml; WAK-Chemie, Steinbach, Germany), incubation at 37°C for 15 to 20 min, and subsequent sonication (6 × 30 s; Branson sonifier W185, level 4). Cell debris was removed by centrifugation (13,000 rpm, 1 min), and the lysate was used for the following reaction. Four hundred microliters of lysate was incubated with 100 μl d-serine (10 mg/ml) for 60 min at 37°C. 2,4-Dinitrophenylhydrazine (Acros; 500 μl; 0.3 mg/ml in 1 N NaCl) was then added, and incubation continued at room temperature for 20 min. The reaction was stopped by addition of 1 ml 2.8 N NaOH. The optical density was read at 450 nm in a spectrophotometer against an assay blank to which all components had been added except d-serine. Additionally, a standard curve with pyruvate from 0.5 μg/ml to 10 μg/ml was measured. Analysis of pyruvate generation was done per mg total protein, which was determined by the method of Markwell (46).

RNA preparation from Staphylococcus saprophyticus.For RNA preparation, wild-type S. saprophyticus or the ΔdsdA mutant was grown in the chemically defined medium supplemented with glucose or glucose and d-serine until an OD₆₀₀ between 0.450 and 0.500 was reached. For each strain and growth condition, RNA was extracted from three biological replicates. Before harvesting, 4 ml Bacteria Protect reagent (Qiagen) was added to 2 ml bacterial culture, and the mixture was incubated for 5 min at room temperature. By a subsequent centrifugation (5,000 rpm, 10 min, room temperature), bacteria were pelleted, and pellets were stored for not longer than 2 weeks at −20°C. RNA was prepared using the RNeasy minikit (Qiagen). The pelleted bacteria were resuspended in 100 μl TE buffer (30 mM Tris, 1 mM EDTA [pH 8.0]), and 10 μl lysostaphin (5 mg/ml) and 10 μl proteinase K (Applichem) were added. After undergoing vortexing and incubation (10 min, 37°C, 100 rpm), bacteria were disrupted mechanically following the manufacturer's instructions with the modification that a Power Vortex was used instead of a TissueLyser. DNase digestion was done on-column using the Qiagen RNase-free DNase set following the manufacturer's instructions. After elution, aliquots of RNA were frozen in liquid nitrogen and stored at −80°C. The integrity and purity of RNA were checked by gel electrophoresis, and absorbance was measured using a Nanodrop 1000. The ratio of A₂₆₀ to A₂₈₀ was between 1.9 and 2.1 for all RNA samples used.

RT-qPCR.Real-time reverse transcription-quantitative PCR (RT-qPCR) was carried out by using the QuantiFast SYBR green RT-PCR kit (Qiagen) for one-step real-time PCR following the manufacturers' instructions and the LightCycler 1.2 (Roche). The reaction volume was 20 μl. RNA was used as the template at a final concentration of 2.5 ng/reaction. Primers were added to a final concentration of 0.5 pmol/each. After RT for 10 min at 50°C, the following temperature protocol was used: an initial activation step for 5 min at 95°C (temperature transition, 20°C/s), followed by a two-step cycling PCR (40 cycles) consisting of denaturation at 95°C for 10 s (temperature transition of 20°C/s), a combined annealing and extension at 60°C for 30 s (temperature transition of 20°C/s), and fluorescence acquisition at 60°C in single mode. Melting curve analysis was performed at 60°C to 95°C (temperature transition, 0.1°C/s). The specificity of the products was checked by melting point analysis and gel electrophoresis.

Calculation of virulence factor expression.For each strain and growth condition, RNA was extracted from three biological replicates. Each replicate was analyzed in triplicate. The test genes where either normalized to gyrB or to dpol. Calculations were done by using the formula 2^{ΔCT test gene}/2^{ΔCT reference gene}, in which the change in cycle threshold (ΔC_T) is the difference of the C_T under glucose conditions − the C_T under d-serine conditions. Values of >2.0 were defined as upregulated, and values of <0.5 were defined as downregulated if observed in all three independent biological replicates.