Isolation and identification of Staphylococcus aureus clinical strains

The 12 S. aureus strains under study were isolated between 1997 and 1998 from infected bone chips (6-10 mm², approximately) of adult patients during surgical debridement treatment of tibiae and femurs in the Bone Infection Service at the, then National Institute of Orthopedics, now the National Institute of Rehabilitation in Mexico City.

Bone specimens were washed twice and suspended in 1.0 ml of phosphate- buffered saline, grinded for one minute in a tissue grinder, and centrifuged at 4000g. Supernatants were streak-seeded in blood agar media, mannitol salt phenol red agar (Merck), and Baird-Parker agar (Merck), and incubated aerobically at 37 °C for 24 h. Characteristic bacterial isolates were identified using the standard biochemical tests and antibiotic susceptibility for S. aureus (Uniscept 20GP System). Gram staining, coagulase (Bactident Coagulase, Merck), and catalase (Catalase Assay Kit, Merck) assays were performed. Clinical strains were methicillinsensitive and were grouped according to infection description: chronic osteomyelitis (17 CPJ, 69 GGT, and 92 HM); infected pseudarthrosis (54 SL, 68 FFC, 88 VTM, 89 RTC, and 93 EMC); and chronic bacterial osteitis (76 IQM, 101 AOC, 105 IMO, and 107 FMR).

Clinical and reference strains (S. aureus ATCC-6538 and S. epidermidis ATCC12228) were preserved in casein peptone-soymeal peptone broth (CASO, Merck) in a solution with 10 % of glycerol at -80 °C.

Bacterial DNA extraction

Genomic DNA was extracted as described by Novick (9) with modifications. Bacterial suspensions grown in brain-heart infusion medium (value of absorbance at 620 nm = 0.8 ) were centrifuged for five minutes at 10,000g; bacterial pellets were resuspended in 20 pl of a mixture of lysostaphin-lysozyme/ Tris-EDTA buffer (TE) at a final concentration of 5.0 pg/pl each (Sigma Chem). After 30 minutes at 37 °C, DNA was extracted using the Genomic DNA Purification Kit (Promega) and resuspended in 50 pl of TE buffer; purity was tested by agarose gel electrophoresis and concentration by ultraviolet (UV) spectroscopy. Extracted DNA was stored in 20 pl aliquots at -20 °C.

SapS gene molecular detection

SapS gene (accession number AY061973) was detected in genomic DNA extracts by end-point polymerase chain reaction (PCR) with the oligonucleotide primers SapS-forward (FW) 5’ - GGCATGAATAAAATTTCAAAG 3’ and SapS-reverse (RV) 5’ GGCTGCAGTTATTTAACTTCGCCTGT - 3’, as previously reported (5). The expected fragment size was 891 base pairs (bp) in the location 1-891.

We also amplified the nucleotide sequence encoding the mature extracellular peptide with acid phosphatase activity using oligonucleotide primers designed by us with the Primer3 software, named SapS2- FW 5’ CCAAAAGTTCTGCTGAAGTTC - 3’ and SapS2-RV 5’ TTATTTAACTTCGCCTGTTTT 3>. The expected fragment size was 800 bp in the location 92-892.

The reaction mixture (20 pl) consisted of 30 ng of target DNA, 200 pM of deoxynucleotide triphosphates (dNTP) (100 mM dNTP set, Invitrogen), 0.5 pM of each oligonucleotide primer pair, 2.0 mM of MgCl₂, and 1.0 unit of thermostable DNA polymerase (Platinum Taq DNA Polymerase, Invitrogen). Amplifications were performed in a thermal cycler (Eppendorf Mastercycler gradient) programmed as follows: an initial 3-minute cycle at 94 °C, 35 cycles of 1-minute steps at 94 °C, 52 °C and 72 °C, and a final 5-minute cycle at 72 °C. After the agarose gel electrophoresis, PCR products were analyzed under UV transilluminator device.

SapS gene detection in Staphylococcus aureus and coagulase-negative staphylococci strains from the database of complete bacterial genomes

An in-silico PCR was performed with the DNA sequences of 49 S. aureus and 11 coagulasenegative Staphylococcus strains found in the database, with in silico analysis of complete bacterial genomes, reported by Bikandi and peers (10). We used the same primer pairs described above. However, given the requirements to fill out the FASTA format for the SapS gene, restriction sites for both forward (FW: 5’ GGC 3’) and reverse (RV: 5’ GGCTGCA 3’) primer sequences were deleted. Thus, the trimmed primers used for in silico SapS amplification were FW: 5’ - ATGAATAAAATTTCAAAG - 3’ and RV: 5’ - GTTATTTAACTTCGCCTGT - 3’.

Experiments were designed to allow two nucleotide mismatches at the 5’ end, no mismatches at the 3’ end, and a maximum number of 3,000 nucleotides for the amplified fragments. Results were obtained 60 seconds after the starting program. A screen capture was made showing fragment amplifications with their molecular sizes.

Sequencing of PCR products and molecular analysis of SapS gene

Sequencing was performed with an Abi Prism 310 sequencer (Genetic Analyzer, Applied Biosystems) using a commercial reagent kit (BigDye Terminator v3.1 Cycle Sequencing Kit, Applied Biosystems). We used 120 ng of each PCR product as DNA template, and 0.5 pM of the SapS forward primer. Electropherograms and nucleotide sequences were stored and analyzed using the Chromas V2 program.

A comparative multiple nucleotide sequence paired global alignment was performed among the nucleotide sequences, and also with the SapS coding sequence (GenBank® AY061973) (5), using the Clustal Omega program ¹¹. The nucleotide sequences from S. aureus clinical and reference strains were translated using the in silico web page ³; the amino acid sequences of the SapS protein were used for analyzing the N-terminal signal peptide domain, its cleavage sequence, and its type of protein secretion system category using the SignalP 5.0 server platform ¹². The subcellular location of SapS for Gram-positive bacteria was analyzed on the Gpos-mPloc platform ¹³. The bipartite signature of aspartate (DDDD) motifs of class C non-specific acid phosphatases was searched using BLAST tools ¹⁴.

Isolation and concentration of protein extracts from bacterial culture media

All bacterial strains were grown in a casein peptone-soymeal peptone broth (CASO, Merck) for 18 h at 37 °C under constant shaking. Bacteria were then separated by centrifugation at 10,000g for 10 minutes at 4 °C; supernatants were sterilized by filtration through 0.22 pm membranes (S-Pak Millipore), aliquoted, and stored at -80 °C. Supernatant (or uninoculated CASO culture medium) protein extracts were obtained by mixing one volume of each supernatant with four volumes of cold (-20 °C) ultrapure acetone (Merck EM Science); the protein was allowed to precipitate at -75 °C for one hour, recovered by centrifugation at 14,000g for 15 minutes at 4 °C, dried at 4 °C, and resuspended in 1.0 ml of 0.1 M sodium acetate solution, pH 5.0 (buffer A). Aliquots (200 pl) were stored at -80 °C. Protein was measured using detergent compatible colorimetric technique (DC Protein Assay, BioRad).

Acid phosphatase activity

Acid phosphatase activity in bacterial culture media was measured as described by Golovan et al.¹⁵ with p-nitro-phenyl phosphate (p-NPP disodium salt, hexahydrate, Sigma Chem). Reaction mixtures (100 pl) containing 10 or 20 pl of the protein extracts and 200 pl of 10 mM p-NPP in buffer A were incubated for 30 minutes at 37 °C; reactions were stopped with 100 pl of 1.0 M NaOH solution, and the p-nitrophenol (p-NP) formed was measured by absorbance at 405 nm (Beckman DU 800 spectrophotometer). Specific acid phosphatase activity was defined as pmoles p-NP/min/mg protein. All measurements were made in triplicate.

To 10 pl of protein extracts, we added one of the following phosphatase inhibitors: 20 mM dibasic dihydrate sodium tartrate, 10 mM sodium fluoride, 2 mM ammonium molybdate tetrahydrate, and 10 mM sodium ortho-vanadate dissolved in buffer A. Reaction mixtures were adjusted to a final volume of 100 pl and incubated initially at 37 °C for 15 minutes. Then, 200 pl of 10 mM p-NPP were added and reaction mixtures were further incubated for 30 minutes at 37 °C. The reaction was stopped with 100 pl of 1.0 M NaOH solution. Specific acid phosphatase activity was calculated as mentioned above.

Identification of extracellular SapS in protein extracts of clinical and reference strains culture media supernatants

SapS presence in protein extracts from culture media supernatants was analyzed by zymography. We seeded 10 pl per well of each clinical and reference protein extract, the noninoculated culture medium together and a standard pre-stained protein molecular weight marker (Precision Plus Protein, All Blue Prestained Protein Standards, BioRad). Protein extracts were run in a 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDSPAGE) under nonreducing conditions, basically as described by Hamilton et al.¹⁶.

After electrophoresis, SDS was removed by washing gels twice with deionized water and once with 1.0% Triton X-100 solution. Then, gels were incubated in renaturing buffer (100 mM Tris HCl, 2 mM MgSO4^6H₂O, 1% Triton X-100 pH 7.0) for 15 minutes at room temperature and equilibrated in buffer A. Acid phosphatase activity was assessed by adding 0.1 % of a-naphthyl phosphate and 0.2% of Fast Garnet GBC (Sigma. Chem) to the buffer and further incubating at 37 °C for 45 minutes. a-naphthol appeared as a brownish-red precipitate on the protein band with phosphatase activity. The reaction was stopped by washing the gel twice with deionized water. Protein bands on parallel gels, run under similar conditions, were stained with Coomassie Brilliant Blue R-250 (BioRad).

SapS specificity for phospho-L-amino-acids and the effect of phosphatase inhibitors

SapS substrate specificity for O-phospho-aminoacids was also determined by zymography according to Slotnick and Gottlieb ¹⁷, with modifications. After SDS-PAGE and renaturation as described above, gel were incubated at 37°C for 45 minutes in 10 ml of Buffer A solution with 4 mM of each of the following substrates: O-phospho-L-serine, O-phospho-L-threonine, and OphosphoLtyrosine (Sigma Chem). Then, gels were washed twice with deionized water. The released inorganic phosphate was detected by gel incubation with a mixture of 0.045% malachite green (oxalate salt, Sigma Chem) and 4.2% ammonium molybdate for 10 minutes at room temperature. The reaction was stopped with two washes of deionized water followed by 5% of acetic acid solution. A positive phosphatase reaction was a green precipitate on the SapS protein band stained with Coomassie Brilliant Blue in a parallel gel, run under similar conditions.

Similarly, phosphatase inhibitors for SapS catalysis were tested by zymography. Renatured gels were incubated at 37 °C for 30 minutes in 10 ml of a 10 mM sodium tartrate, sodium fluoride, sodium ortho-vanadate, or ammonium molybdate (Sigma Chem) solution. Gels were washed once in buffer A and further incubated at 37 °C for another 30 minutes in 10 ml of 0.1% a-naphthyl phosphate or 4 mM of each of the three O-phospho-amino acids as substrates. The corresponding reaction products, a-naphthol and released phosphate, were stained as described above with 0.2% Fast Garnet GBC or malachite green/ammonium molybdate, respectively.

Statistical analysis

The statistical significance of differences among the means of acid phosphatase measurements with and without enzyme inhibitors was determined with an ANOVA test and a comparative Tukey post-hoc test using the GraphPad Prism 5.0 software. Significance was set at p<0.05.

SapS gene is present in the genomes of isolated S. aureus clinical strains from chronically osteomyelitic patients

Using the primers designed for S. aureus 154 ⁸, we detected the presence of the complete SapS gene as a single 891 bp-long fragment in the 12 clinical strains and S. aureus ATCC-6538, but not in S. epidermidis ATCC-12228. With SapS2 primer set we obtained a single 805 bp-long fragment that matched the nucleotide sequence of the extracellular mature peptide with acid phosphatase activity. These results show that the SapS gene and the nucleotide sequence encoding the extracellular phosphatase are present in the clinical and the reference S. aureus strains, but not in S. epidermidis ATCC-12228.

SapS gene is present in S. aureus strains but not in the coagulasenegative Staphylococcus strains from the database with complete bacterial genomes.

In silico PCR confirmed the presence of the SapS gene only in the S. aureus species. Using the original primer set ⁸, the 892 bp-long fragment was obtained for all the 49 S. aureus strains tested with the in silico PCR (figure 1A) and none for any of the 11 coagulase-negative Staphylococcus strains. Similarly, with the SapS2 primers, only a single 800 bp-fragment was obtained for S. aureus strains (figure 1B); no amplifications were obtained with either primers for any of the plasmids contained in the complete bacterial genomes database.

Figure 1 In silico PCR amplifications. A. SapS gene amplifications using the oligonucleotide primers described by du Plessis, et al. ⁸. B. Amplifications of the SapS gene region encoding for the mature extracellular acid phosphatase (SapS2) using our set of oligonucleotide primers. In both cases, DNA templates were from the 49 S. aureus strains and the 11 coagulase-negative Staphylococcus strains, and from the plasmids housed in the database with the in silico analysis of bacterial genomes ¹⁰. Amplifications of six coagulase-positive S. aureus strains are shown as examples. No amplifications were found for coagulase-negative strains. 

The SapS gene nucleotide and amino acid sequences reveal the presence of the protein SapS catalytic domain and the signal peptide cleavage site.

The multiple nucleotide sequence analysis performed with the SapS sequences of S. aureus 154 ⁸, the 12 clinical S. aureus strains, and the S. aureus ATCC-6538 reference strain showed an identity percentage of 92-96% (table 1) suggesting that SapS gene from the assessed S. aureus strains are homologous. The built phylogenetic tree (data not shown) revealed the closeness among the sequences implying they could share a common ancestor.

The amino acid sequence analysis of the SapS gene on the SignalP 5.0 server showed the presence of the signal peptide in the N-terminal amino acid sequence (27 - STAFAKSSAEVQQ - 39) in the 12 clinical S. aureus strains (table 2) as previously reported ⁷. It included the recognition site for signal peptide cleavage between residues alanine and lysine in the positions 31 and 32 (AK) respectively.

The predictive analysis on the Gpos-mLoc server suggested that SapS is a typical extracellular protein belonging to the S. aureus secretome ¹⁸. Its bipartite motifs A (101 - ALDLDETVLDNSPY - 114) and B (213 - LVMLFGDNLLDF - 224), and the associated four invariant residues of aspartate ¹⁹ - with dephosphorylating enzymatic activity of class “C” NSAPs ²⁰ - were also found to be highly conserved in our 12 clinical S. aureus strains (table 2).

SapS resistance to sodium tartrate and sodium fluoride, and sensitivity to sodium ortho-vanadate and ammonium molybdate

The average acid phosphatase-specific activity was 25.35 mmoles of p-nitrophenol per minute and per mg of protein at pH 5.0 (figure 2) in the extracts from the culture media of the 12 S. aureus clinical strains incubated with p-nitro-phenyl phosphate as substrate. Staphylococcus aureus ATCC- 6538 showed the lowest level of activity (4.5 mmoles p-NP/min/mg of protein) and S. epidermidis ATCC-12228 strain showed no activity at all. Moreover, this extracellular acid phosphatase activity was resistant to inhibition by both sodium tartrate, a general phosphatase inhibitor, and sodium fluoride, an inhibitor of acid phosphatases (figure 3). However, this activity was inhibited by sodium ortho-vanadate (~ 50%) and ammonium molybdate (80-95%) (figure 2).

Figure 2 Percentageof appS activityinthepresenceandabsenceof phosphatasesinhibitors. SppS activitywasassessed from protein extracts of culture media of clinical strains; 10 mM ofp-nitro-phenyl-phosphate was used as substrate. Assays were run in triplicate from three independent experiments; results are expressed as mean ± standard error of the mean. Statistical analysis was performed using ANOVA and significance was set at *p<0.05. 

Figure 3 Non-reducing gel electrophoresis and zymographic analysis of protein extracts obtained from the culture media of clinical bacterial strains, reference strains, and the sterile culture medium (SCM). A. Coomassie Brilliant Blue for protein staining. B. Brown red a-naphthol-Fast Garnet GBC complex for SapS activity assessment, and C. Phosphate/green malachite/ ammonium molybdate complex for SapS phosphatase activity on O-phospho- L-tyrosine. Lanes from left to right: 1, pre-stained protein molecular weight; 2 13, clinical bacterial strains; 17, CPJ, 54 SL, 68 FFC, 69 GGT, 76 IQM, 88 VTM, 89-RTC, 92-HM, 93-EMC,101-AOC,105-IMO,107-FMR; 14, Stpphylococcus pureus ATCC-6538 strain; 15, Stpphylococcus epidermidis ATCC-12228; 16, SCM. 

SapS is present in semi-purified culture media supernatant protein extracts of S. aureus clinical strains.

Non-reducing SDS-PAGE of partially purified culture media supernatant protein extracts of S. pureus clinical strains and S. pureus ATTC-6538 (figure 3A) showed remarkably similar extracellular protein profiles (Coomassie brilliant blue-staining) ranging from 30 to 100 kDa. Zymograms analysis showed only one extracellular protein band with phosphatase activity for clinical S. pureus and S. pureus ATCC-6538 strains, as revealed by the appearance of a-naphthol at the same migration distance as the protein band run in a parallel gel, stained with Coomassie brilliant blue. Both bands showed an approximate molecular mass of 30 kDa (figures 3A and 3B) strongly suggesting their identity with SapS. The protein extract from the S. epidermidis ATCC-12228 (figures 3A and 3B, lane 15) showed no 30 kDa- protein band, nor associated enzymatic activity.

SapS exhibits O-phospho-L-tyrosine phosphatase activity and is sensitive to sodium vanadate and ammonium molybdate.

The zymograms analysis of the culture media protein extracts of all S. aureus clinical strains and S. aureus ATTC-6538 strain in the presence of Ophospho-amino-acids showed a green malachite/molybdate/phosphate complex precipitate at the same migration distance as the 30 kDa observed in the Coomassie stained-gel and the brownish-red a-naphthol/Fast Garnet GBC precipitate staining. These results confirmed SapS phosphatase activity (figure 3C).

However, SapS showed O-phospho-L-tyrosine phosphatase activity while neither OphosphoLserine, nor O-phospho-L-threonine was dephosphorylated. It appears, therefore, that SapS and the o-phospho-L-tyrosine phosphatase are the same enzyme.

Furthermore, in another set of parallel zymograms including one of the four phosphatase inhibitors mentioned, we found that the green malachite/ molybdate/phosphate complex precipitate, formed after O-phospho-tyrosine dephosphorylation, was resistant to sodium tartrate and sodium fluoride, but did not appear in the presence of sodium vanadate and ammonium molybdate.